System and method for monitoring and controlling biological production process by mid-infrared spectroscopy

Abstract

Methods and systems for monitoring and / or controlling production units used at various stages of biological product development and manufacturing are presented herein. In particular, in certain embodiments, the biological production monitoring and control techniques described herein utilize a mid-infrared (mid-IR) analyzer that is capable of obtaining mid-infrared spectral data from an aqueous sample in substantially real time. The techniques described herein can utilize this mid-infrared spectroscopic data to measure sample quality metrics, such as protein content, titer, secondary structure, aggregation, and the like, as well as viral and / or nucleic acid characteristics. Thus, sample quality metrics can be measured substantially in real time and / or continuously to assess, for example, the production quality of protein and gene therapeutic agents, and can be used by a control system to adjust and / or control process parameters to improve product quality, etc.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims the priority and benefit of U.S. Provisional Application No. 63 / 425,504, filed on November 15, 2022, U.S. Provisional Application No. 63 / 431,989, filed on December 12, 2022, and U.S. Provisional Application No. 63 / 438,969, filed on January 13, 2023, the contents of each of which are hereby incorporated by reference in their entirety. BACKGROUND OF THE INVENTION

[0003] Biological products (also referred to as biologic drugs) are a rapidly growing and increasingly important class of pharmaceutical products obtained (e.g., isolated) from natural sources such as humans, animals, or microorganisms. They can include a range of products such as vaccines, blood and blood components, allergens, somatic cells, gene therapy components, tissues, and proteins. For example, recently approved recombinant protein therapeutics have been developed to treat a variety of clinical indications, including cancer, autoimmune / inflammatory, exposure to infectious agents, and genetic disorders. Gene therapy has great potential to transform the lives of patients with genetic diseases.

[0004] Unlike conventional small molecule drugs that are chemically synthesized and have known structures, biologic products are manufactured through highly complex biological processes and are complex mixtures as the final product or at various steps along the production line, which are difficult to identify and / or characterize. Multiple-step purification methods are typically required to obtain a consistent, pure, and effective final product.

[0005] Accordingly, the difficulties associated with accurately identifying and characterizing biological products and their components during the manufacturing cycle pose obstacles to developing and testing the manufacturing processes of new biologic drugs, scaling up production capacity after approval and / or responding to changes in demand, and improving and controlling existing procedures to ensure consistent product quality, avoid adverse events, and reduce downtime. Thus, there is a need for improved techniques for monitoring the inputs and outputs of biological product processing steps and manufacturing controls. SUMMARY OF THE INVENTION

[0006] The present disclosure provides methods and systems for monitoring and / or controlling production units used at various stages of bioproduct development and manufacturing. In particular, in certain embodiments, the bioproduction monitoring and control techniques described herein utilize mid-infrared (mid-IR) analyzers that are capable of obtaining mid-IR spectral data from aqueous samples substantially in real time. The techniques described herein can utilize this mid-IR spectral data to measure sample quality metrics such as protein content, titer, secondary structure, aggregation, etc., as well as virus and / or nucleic acid characteristics such as empty / full capsid measurements and other sample characteristics that, in certain embodiments, can be used as critical quality attributes (CQAs) in accordance with Food and Drug Administration (FDA) drug development guidelines. Accordingly, sample quality metrics can be measured substantially in real time and / or continuously to continuously evaluate, for example, the production quality of protein therapeutics and gene therapeutics.

[0007] The control system and method can in turn use this real-time data to control and / or adjust process parameters such as collection windows, flow rates, salt gradients, etc. for collecting target elution fractions during a chromatographic elution process to improve target recovery, sample purity, potency, stability, etc.

[0008] In this manner, by facilitating control and improvement of downstream sample processing, the techniques described herein allow for more efficient and robust processing, with improved results and reduced variability in product quality (e.g., purity and potency). These translate to increased success rates in FDA trials, faster time to market, facilitated scale-up, and reduced costs, ultimately providing more effective and accessible therapeutics to patients in need.

[0009] In one aspect, the present disclosure relates to a method for obtaining a purified sample of a target protein species by real-time monitoring of protein heterogeneity and (e.g., automated; e.g., semi-automated) control of a purification process, the method comprising: (a) at each of one or more time points, measuring, by one or more mid-infrared (MIR) analyzers, a corresponding infrared (IR) absorbance signal of an aqueous sample exiting a purification unit (e.g., a chromatography column), the aqueous sample comprising one or more protein species including the target protein species; (b) receiving, by a processor of a computing device, IR absorbance data corresponding to the IR absorbance signal at each of the one or more time points; (c) determining, by the processor, values of one or more sample quality metrics based on the IR absorbance data, the one or more sample quality metrics including a protein aggregation metric indicative of a level of protein aggregation within the aqueous sample; and (d) using the one or more sample quality metrics to control collection of a target fraction of the aqueous sample (e.g., during a specific collection window) to thereby obtain the purified sample of the target protein species.

[0010] In certain embodiments, the purification unit is or includes a chromatography column. In certain embodiments, the chromatography column is a member selected from the group consisting of: an affinity chromatography (AC) column, a hydrophobic interaction chromatography (HIC) column, an ion exchange chromatography (IEX) column, a size exclusion chromatography (SEC) column, and a mixed-mode chromatography column [e.g., any combination of the foregoing chromatography columns (e.g., IEX and HIC; e.g., IEX and SEC)].

[0011] In certain embodiments, the target protein species is or includes one or more members selected from the group consisting of: monoclonal antibody (mAb), fusion protein, virus capsid protein, antibody-drug conjugate, recombinant protein, and plasma protein. In certain embodiments, the target protein species is or includes a peptide chain and / or a protein fragment.

[0012] In certain embodiments, the aqueous sample includes a plurality of different protein species. In certain embodiments, the aqueous sample includes a heterogeneous population of the target protein species, which includes a monomeric portion and an aggregated portion. In certain embodiments, the target protein species is a subspecies of a specific protein species, and the target protein species has a specific desired level and / or type of molecular conjugation (e.g., glycan, small molecule drug, polyethylene glycol, etc.).

[0013] In certain embodiments, at least one of the one or more MIR analyzers is or includes a MIR spectrometer, the MIR spectrometer including: a MIR source that is aligned and operable to emit a MIR beam [e.g., including a wavelength range that is substantially within the MIR spectral range (e.g., a range of about 5000 cm -1 to about 500 cm -1 (e.g., about 2 to 20 microns))]; one or more sampling optics that are aligned to direct and / or allow the MIR beam and / or at least a portion thereof to pass through and / or contact at least a portion of the aqueous sample [e.g., wherein the MIR beam contacts the portion of the aqueous sample by reflection at an interface between a solid material (e.g., ATR crystal and / or optical fiber) and the aqueous sample (e.g., wherein the MIR beam undergoes total internal reflection and contacts / detects the portion of the aqueous sample through an evanescent wave extending into the aqueous sample)], and after passing through or contacting the portion of the aqueous sample, towards one or more detectors; and the one or more detectors that are aligned and operable to detect the MIR beam after the MIR beam passes through and / or contacts the aqueous sample.

[0014] In certain embodiments, one or more sampling optics include a high refractive index material (e.g., an ATR crystal; e.g., an optical fiber) that is aligned such that the MIR beam is directed towards an interface between the high refractive index material and an aqueous sample, incident on the interface, and internally reflected by the interface (e.g., back into the high refractive index material) (e.g., such that the MIR beam is incident on the interface at an angle greater than the critical angle for total internal reflection); and one or more detectors are aligned and operable to detect the MIR beam that exits the high refractive index material after being internally reflected by the high refractive index material. In certain embodiments, the high refractive index material is an ATR crystal. In certain embodiments, the high refractive index material is an optical fiber.

[0015] In certain embodiments, one or more sampling optics include a flow cell that includes a detection channel through which an aqueous sample flows; and one or more detectors are aligned and operable to detect the MIR beam that exits the detection channel after the MIR beam has transmitted through the detection channel. In certain embodiments, the path length through the detection channel (e.g., the path followed by the MIR beam during transmission) is about 10 μm or greater (e.g., about or at least 15 μm or greater; e.g., about 25 μm or greater; e.g., about 30 μm or greater; e.g., about 40 μm or greater; e.g., about 50 μm or greater).

[0016] In certain embodiments, one or more MIR analyzers are or include a quantum cascade laser (QCL)-based MIR spectrometer that includes a QCL source operable to emit an MIR beam [e.g., including a wavelength range that is substantially within the MIR spectral range (e.g., a range from about 5000 cm -1 to about 500 cm -1 (e.g., from about 2 to 20 microns))].

[0017] In certain embodiments, (e.g., the MIR source is a laser, and) the MIR beam has a spectral linewidth of about 4 cm -1 or less (e.g., about 2 cm -1 or less; e.g., about 1 cm -1 or less; e.g., about 0.5 cm -1 or less).

[0018] In certain embodiments, the power of the MIR beam is about 1 mW or greater (e.g., about 10 mW; e.g., about 50 mW or greater; e.g., about 100 mW or greater; e.g., about 500 mW or greater; e.g., about 1000 mW or greater).

[0019] In certain embodiments, the spectral resolution of the MIR spectrometer is about 4 cm -1or better (e.g., lower); e.g., about 2 cm -1 or better (e.g., lower); e.g., about 1 cm -1 or better (e.g., lower); e.g., about 0.5 cm -1 or better (e.g., lower); e.g., about 0.25 cm -1 or better (e.g., lower); e.g., about 0.1 cm -1 or better (e.g., lower); e.g., about 0.05 cm -1 or better (e.g., lower).

[0020] In some embodiments, the (e.g., frequency / wavelength) accuracy of the MIR spectrometer is about 2 cm -1 or better (e.g., lower); e.g., about 1 cm -1 or better (e.g., lower); e.g., about 0.5 cm -1 or better (e.g., lower); e.g., about 0.25 cm -1 or better (e.g., lower); e.g., about 0.1 cm -1 or better (e.g., lower); e.g., about 0.01 cm -1 or better (e.g., lower).

[0021] In some embodiments, the (e.g., frequency / wavelength) repeatability of the MIR spectrometer is about 0.5 cm -1 or better (e.g., lower); e.g., about 0.25 cm -1 or better (e.g., lower); e.g., about 0.1 cm -1 or better (e.g., lower); e.g., about 0.05 cm -1 or better (e.g., lower); e.g., about 0.001 cm -1 or better (e.g., lower).

[0022] In certain embodiments, the MIR source is a tunable laser (e.g., a tunable QCL) (e.g., operable to scan the emission frequency of the MIR beam over a plurality of frequencies within a scan range), and the method includes, for each of one or more time points: scanning the emission frequency of the MIR beam within the scan range of the tunable laser so as to irradiate an aqueous sample at a plurality of emission frequencies; and detecting the MIR beam (e.g., which has (i) been internally reflected by an interface between a high refractive index material and the aqueous sample and / or (ii) transmitted through a detection channel through which the aqueous sample flows) at each of the plurality of emission frequencies with one or more detectors so as to measure a corresponding infrared (IR) spectrum including a plurality of values as a corresponding IR absorbance signal of the aqueous sample, each of the plurality of values being associated with and representing and / or based on the power detected at a particular one of the plurality of emission frequencies.

[0023] In certain embodiments, the plurality of emission wavelengths include one or more wavelengths within a spectral band in the range of from about 1800 to about 800 cm -1 (e.g., from about 1725 to about 1025 cm -1 ; e.g., from about 1750 to about 1350 cm -1 ; e.g., from about 1725 to about 1375 cm -1 ; e.g., from about 1700 to about 1500 cm -1 ; e.g., from about 1700 to about 1600 cm -1 ; e.g., from about 1700 to about 1000 cm -1 ). In certain embodiments, the plurality of emission wavelengths include one or more wavelengths within a spectral band in the range of from about 1600 cm -1 to about 1500 cm -1 .

[0024] In certain embodiments, the MIR analyzer is an in-line sensor (e.g., as opposed to an off-line or at-line sensor) that measures the IR absorbance signal substantially in real time as the aqueous solution exits a purification unit.

[0025] In certain embodiments, for each of one or more time points, the IR absorbance data includes a corresponding amide band spectrum [e.g., for each particular wavelength of a plurality of sampled (e.g., emitted) wavelengths in the range of from about 1800 to about 800 cm -1 , the amide band spectrum including a relevant absorption value representing the absorption level of a portion of the aqueous sample at the particular wavelength].

[0026] In certain embodiments, step (d) includes determining a corresponding value of a protein aggregation metric for each particular time point for at least a portion of the one or more time points.

[0027] In some embodiments, for each particular time point, determining the corresponding value of the protein aggregation metric includes: calculating a value of an amide II peak metric from an amide band spectrum corresponding to the particular time point, the amide II peak metric quantifying one or more characteristics (e.g., frequency position, line width, intensity) of the amide II band at the particular time point; and using the value of the amide II peak metric to determine the corresponding value of the protein aggregation metric (e.g., where the protein aggregation metric is the amide II peak metric or a function of the amide II peak metric).

[0028] In some embodiments, the amide II peak metric is a frequency position metric that quantifies the frequency at which the amide II band is substantially centered at the particular time point [e.g., the centroid frequency, the frequency of the maximum height of the amide II band, the center frequency of a fitted peak function (e.g., Gaussian, Lorentzian, etc.)]. In some embodiments, the frequency position metric is the centroid frequency of the amide II band.

[0029] In some embodiments, for each particular time point, determining the corresponding value of the protein aggregation metric includes: calculating a value of an amide I peak metric from an amide band spectrum corresponding to the particular time point, the amide I peak metric quantifying one or more characteristics (e.g., frequency position, line width, intensity) of the amide I band at the particular time point; and using both the value of the amide I peak metric and the value of the amide II peak metric to determine the corresponding value of the protein aggregation metric (e.g., where the protein aggregation metric is a function of the amide I peak metric and the amide II peak metric).

[0030] In some embodiments, the amide I peak metric is a peak intensity metric that quantifies the intensity of the amide I band at the particular time point (e.g., the peak height of the amide I band, the area under the curve (AUC) of the amide I band); the amide II peak metric is a peak intensity metric that quantifies the intensity of the amide II band at the particular time point (e.g., the peak height of the amide I band, the area under the curve (AUC) of the amide I band); and determining the value of the protein aggregation metric includes calculating (i) the ratio of the value of the amide I peak metric to the value of the amide II peak metric and / or (ii) the ratio of the value of the amide II peak metric to the value of the amide I peak metric.

[0031] In some embodiments, the one or more sample quality metrics further include a total protein content metric that indicates the level of protein content within the aqueous sample.

[0032] In some embodiments, step (d) includes the processor causing one or more trigger signals (e.g., voltages) to be transmitted to a controller unit of the purification unit. In some embodiments, the one or more trigger signals include an analog voltage signal having a time-varying amplitude based on a value of a protein aggregation metric (e.g., substantially proportional thereto). In some embodiments, the one or more trigger signals include an analog voltage signal having a time-varying amplitude based on a value of a total protein content metric (e.g., substantially proportional thereto).

[0033] In some embodiments, step (d) includes one or both of the following: initiating collection of a target fraction of the aqueous sample by the controller unit based on one or more trigger signals [e.g., where a particular one of the one or more trigger signals is an analog signal and the controller unit initiates collection of the target fraction based on the amplitude of the analog signal (e.g., when it exceeds a particular threshold; e.g., when it drops below a particular threshold); e.g., where a particular one of the one or more trigger signals is a digital signal that triggers (e.g., by transitioning from a 0 voltage level to a 1 voltage level, or vice versa) initiation of collection of the target fraction] and stopping collection of the target fraction of the aqueous sample by the controller unit based on the one or more trigger signals [e.g., where a particular one of the one or more trigger signals is an analog signal and the controller unit stops collection of the target fraction based on the amplitude of the analog signal (e.g., when it exceeds a particular threshold; e.g., when it drops below a particular threshold); e.g., where a particular one of the one or more trigger signals is a digital signal that triggers (e.g., by transitioning from a 0 voltage level to a 1 voltage level, or vice versa) stopping of collection of the target fraction].

[0034] In another aspect, the present disclosure relates to a method for real-time monitoring of protein aggregation in a sample, the method comprising: (a) repeatedly receiving, by a processor of a computing device, IR absorbance data corresponding to IR absorbance signals measured at each of a plurality of time points, wherein for each particular time point of the plurality of time points, the IR absorbance data comprises a corresponding IR absorbance spectrum measured from the sample at the particular time point and comprises a plurality of absorbance values, each absorbance value being associated with a particular wave number; (b) analyzing, by the processor (e.g., automatically), the IR absorbance data to obtain a real-time protein aggregation signal that provides a measure of protein aggregation in the sample as a function of time, for each particular time point of the plurality of time points: determining values of one or more peak metrics for one or both of the amide I band and the amide II band using the IR absorbance spectrum corresponding to the particular time point; determining a value of a protein aggregation metric indicative of the level of protein aggregation in the sample at the particular time point using the values of the one or more peak metrics; and updating the real-time protein aggregation signal with the determined value of the protein aggregation metric at the particular time point; and (c) storing and / or providing, by the processor, the real-time protein aggregation signal for one or more of the following uses: (i) further processing, (ii) display, and (iii) use as a control signal for adjusting one or more purification units (e.g., a chromatography system).

[0035] In certain embodiments, step (b) comprises, for each particular time point, determining a value of a frequency position metric as the value of a protein aggregation metric indicative of the level of protein aggregation in the sample at the particular time point, the frequency position metric quantifying the frequency at which the amide II band is substantially centered at the particular time point [e.g., the centroid frequency, the frequency of the maximum height of the amide II band, the center frequency of a fitted peak function (e.g., Gaussian, Lorentzian, etc.)]. In certain embodiments, the frequency position metric is the centroid frequency of the amide II band.

[0036] In certain embodiments, for each particular time point, step (b) comprises: determining a value of an amide I peak intensity metric that quantifies the intensity of the amide I band at the particular time point (e.g., the peak height of the amide I band, the area under the curve (AUC) of the amide I band); determining a value of an amide II peak intensity metric that quantifies the intensity of the amide II band at the particular time point (e.g., the peak height of the amide I band, the area under the curve (AUC) of the amide I band); and determining (i) the ratio of the amide I peak metric value to the amide II peak metric value and / or (ii) the ratio of the amide II peak metric value to the amide I peak metric value as the value of the protein aggregation metric.

[0037] In another aspect, the present disclosure relates to a method for monitoring and controlling a production unit for manufacturing biological products (e.g., proteins; e.g., nucleic acids; e.g., viruses) based on mid-infrared (MIR) spectroscopy, the method comprising: (a) measuring, at each of one or more time points, the corresponding infrared (IR) absorbance signals of aqueous samples (e.g., and which include one or more inputs, output products, waste products, or in-process products of the production unit) flowing into and / or out of the production unit by one or more (e.g., integrated) mid-infrared (MIR) analyzers [e.g., the MIR analyzers described in one or more aspects and / or embodiments herein (e.g., in the foregoing paragraphs)]; (b) receiving, by a processor of a computing device, IR absorbance data corresponding to the IR absorbance signals at each of the one or more time points; and (c) using the received IR absorbance data to adjust one or more process parameters of: (i) the production unit and / or (ii) a second (e.g., upstream and / or downstream) production unit.

[0038] In certain embodiments, the production unit is or includes a purification unit. In certain embodiments, the purification unit is a member selected from the group consisting of: an alternating tangential flow filtration (ATF) system, a tangential flow depth filtration (TFDF) system, a tangential flow filtration (TFF) system, a chromatography column, a direct or conventional flow filtration unit, an ultrafiltration unit, and a diafiltration unit. In certain embodiments, the purification unit is a chromatography column {e.g., and wherein the chromatography column is a member selected from the group consisting of: an affinity chromatography (AC) column, a hydrophobic interaction chromatography (HIC) column, an ion exchange chromatography (IEX) column, a size exclusion chromatography (SEC) column, and a mixed-mode chromatography column [e.g., any combination of the foregoing chromatography columns (e.g., IEX and HIC; e.g., IEX and SEC)]}.

[0039] In certain embodiments, the production unit is or includes a bioreactor (e.g., a seed bioreactor; e.g., a production bioreactor).

[0040] In certain embodiments, the aqueous sample includes one or more protein species selected from the group consisting of: monoclonal antibodies (mAbs), fusion proteins, viral capsid proteins, antibody-drug conjugates, recombinant proteins, and plasma proteins. In certain embodiments, the aqueous sample includes peptide chains and / or protein fragments.

[0041] In certain embodiments, the aqueous sample includes a plurality of different protein species. In certain embodiments, the aqueous sample includes a heterogeneous population of a target protein species, which includes monomeric and aggregated portions. In certain embodiments, the aqueous sample includes one or more subspecies of a specific protein species having a specific desired level and / or type of molecular conjugation (e.g., glycan, small molecule drug, polyethylene glycol, etc.).

[0042] In certain embodiments, the aqueous sample comprises nucleic acids (e.g., DNA, RNA, mRNA, etc.).

[0043] In certain embodiments, the aqueous sample comprises one or more viruses and / or virus-like particles [e.g., adeno-associated virus vectors (AAV); e.g., lentiviral vectors].

[0044] In certain embodiments, step (c) comprises using IR absorbance data to determine the value of one or more sample quality metrics at each of one or more time points (e.g., and adjusting one or more process parameters based thereon).

[0045] In certain embodiments, one or more sample quality metrics include a total protein content metric that quantifies the amount and / or concentration of protein within the aqueous sample. In certain embodiments, one or more sample quality metrics include a protein aggregation metric that indicates the level of protein aggregation within the aqueous sample. In certain embodiments, one or more sample quality metrics include one or more protein species metrics that identify the presence and / or quantify the amount of one or more specific protein species within the aqueous sample (e.g., absolute amount; e.g., relative amount). In certain embodiments, one or more sample quality metrics include a protein conjugation metric that quantifies the level and / or type of molecular conjugation (e.g., glycan, small molecule drug, polyethylene glycol, etc.). In certain embodiments, one or more sample quality metrics include one or more protein secondary structure metrics that quantify the presence and / or amount of one or more protein secondary structure motifs (e.g., α-helix content, β-sheet content, turn content, disordered content).

[0046] In certain embodiments, one or more sample quality metrics include one or more nucleic acid content metrics that quantify the nucleic acid content within the aqueous sample [e.g., the total nucleic acid content (e.g., concentration (e.g., titer), mass per volume (e.g., mg / mL, number of particles per volume, viral genome copy number, etc.); e.g., the sum and / or relative amount of one or more specific types of nucleic acids (e.g., DNA, RNA, ssDNA, dsDNA), e.g., measurement of the independent and / or distinguishable content metrics of viral nucleic acid and host cell nucleic acid; e.g., the total amount of a specific nucleobase in a nucleic acid sample, e.g., GC content].

[0047] In certain embodiments, one or more sample quality metrics include one or more virus content metrics that quantify the content of viruses and / or virus-like particles in an aqueous sample [e.g., concentration (e.g., titer), mass per volume (e.g., mg / mL), number of (viral) particles per volume, etc.]. In certain embodiments, one or more sample quality metrics include one or more empty / full capsid ratios that quantify the content and / or relative fraction of empty and / or full virus vectors in an aqueous sample (e.g., percentage of full virus vectors, ratio, etc.). In certain embodiments, one or more sample quality metrics include a capsid aggregation metric that indicates the level of capsid aggregation in a virus vector sample. In certain embodiments, one or more sample quality metrics include a viral nucleic acid (e.g., viral DNA, RNA, etc.) content metric that differentiates viral nucleic acid from host cell protein and host cell nucleic acid content.

[0048] In certain embodiments, at least a portion (e.g., one or more) of the sample quality metrics are calculated based on one or more peak metrics that measure characteristics of one or more absorption bands in IR spectral data {e.g., where each peak metric is associated with one or more specific spectral bands [e.g., a continuous range of wavelengths / wavenumbers (e.g., amide-I band, amide-II band, amide-III band; e.g., the amide region spanning two or more amide bands; e.g., the asymmetric PO4 band; e.g., the symmetric PO4 band)], and quantify specific structural characteristics of one or more absorption peaks within the specific spectral band [e.g., intensity (e.g., peak amplitude; e.g., area under the curve (AUC)); e.g., linewidth; e.g., frequency position (e.g., peak frequency; e.g., centroid frequency)]}.

[0049] In certain embodiments, the IR absorbance data includes: (i) one or more (e.g., multiple) amide II absorbance values that are associated with and measure the IR absorption of a (viral sample) at a wavenumber within the amide II spectral band (e.g., in the range of about 1500 cm -1 to about 1600 cm -1 , e.g., in the range of about 1500 cm -1 to about 1575 cm -1 ; e.g., in the range of about 1500 cm -1 to about 1550 cm -1 ; e.g., in the range of about 1540 cm -1 to about 1560 cm -1 ); and / or (ii) one or more (e.g., multiple) amide III absorbance values that are associated with and measure the IR absorption of a (viral sample) at a wavenumber within the amide II spectral band (e.g., in the range of about 1250 cm -1 to about 1350 cm -1 , e.g., in the range of about 1250 cm -1 to about 1325 cm-1 ; for example, in the range of about 1275 cm -1 to about 1325 cm -1 ; for example, in the range of about 1280 cm -1 to about 1300 cm -1 ) and measuring the IR absorption associated with the (virus sample) at the wavenumber.

[0050] In certain embodiments, determining one or more sample quality metrics includes determining a value of a protein content metric [e.g., concentration (e.g., titer)] at least in part based on amide II and / or amide III absorbance values, the protein content metric quantifying the protein content within the sample. In particular embodiments, the method includes determining a value of an amide II peak metric based on the amide II absorbance value and / or determining a value of an amide III peak metric based on the amide III absorbance value; and using the amide II peak metric value and / or the amide III peak metric value to determine the protein content metric value. In certain embodiments, the amide II peak metric and / or the amide III peak metric are peak intensity metrics that respectively quantify the intensity [e.g., peak height, area under the curve (AUC), etc.] of the amide II band and / or the amide III band.

[0051] In certain embodiments, the IR absorbance data includes: (i) one or more (e.g., multiple) antisymmetric phosphate stretch (antisymmetric-PO4) absorbance values associated with the IR absorption of the (virus sample) within the antisymmetric-PO4 spectral band (e.g., in the range of about 1150 cm -1 to about 1250 cm -1 ; for example, in the range of about 1175 cm -1 to about 1250 cm -1 ; for example, in the range of about 1200 cm -1 to about 1250 cm -1 ; for example, in the range of about 1210 cm -1 to about 1230 cm -1 ) and measuring the IR absorption; and / or (ii) one or more (e.g., multiple) symmetric phosphate stretch (symmetric-PO4) absorbance values associated with the IR absorption within the symmetric-PO4 spectral band (e.g., in the range of about 1000 cm -1 to about 1100 cm -1 ; for example, in the range of about 1050 cm -1 to about 1100 cm -1 ; for example, in the range of about 1075 cm -1 to about 1100 cm -1 ; for example, in the range of about 1075 cm -1 to about 1085 cm -1) associated with the IR absorption (of the viral sample) at the wavenumber and measuring the IR absorption.

[0052] In certain embodiments, determining one or more sample quality metrics includes determining a value of a nucleic acid (e.g., DNA, RNA, etc.) content metric that is at least partially based on the antisymmetric-PO4 and / or symmetric-PO4 absorbance values, the nucleic acid content metric quantifying the nucleic acid content [e.g., concentration (e.g., titer)] within the sample. In certain embodiments, the method includes determining a value of an antisymmetric-PO4 peak metric based on the antisymmetric-PO4 absorbance value and / or determining a value of a symmetric-PO4 peak metric based on the symmetric-PO4 absorbance value; and using the antisymmetric-PO4 peak metric value and / or the symmetric-PO4 peak metric value to determine the nucleic acid content metric value. In certain embodiments, the antisymmetric-PO4 peak metric and / or the symmetric-PO4 peak metric are peak intensity metrics that respectively quantify the intensity [e.g., peak height, area under the curve (AUC), etc.] of the antisymmetric-PO4 band and / or the symmetric-PO4 band.

[0053] In certain embodiments, determining one or more sample quality metrics includes determining (i) a value of a protein content metric that quantifies the protein content [e.g., concentration (e.g., titer)] within the sample and (ii) a value of a nucleic acid (e.g., DNA, RNA, etc.) content metric that quantifies the nucleic acid content [e.g., concentration (e.g., titer)] within the sample, thereby independently quantifying the total protein and nucleic acid content within the sample. In certain embodiments, determining one or more sample quality metrics includes determining the total capsid content at least partially based on the value of the protein content metric (e.g., as a function thereof). In certain embodiments, determining one or more sample quality metrics includes determining the full capsid fraction at least partially based on (i) the value of the protein content metric and / or the total capsid content and (ii) the value of the nucleic acid content metric (e.g., as a function thereof).

[0054] In certain embodiments, the IR absorbance data is or includes one or more IR absorbance spectra, and for each specific wavenumber among a plurality of wavenumbers spanning the measured spectral band, each IR absorbance spectrum includes a corresponding IR absorbance value that represents a measure of the absorption of IR light by the aqueous sample at the specific wavenumber.

[0055] In certain embodiments, the measured spectral band spans one or more bands selected from the group consisting of the amide II band, the amide III band, the asymmetric-PO4 band, and the symmetric-PO4 band.

[0056] In some embodiments, the provided method includes determining values of one or more sample quality metrics at each of one or more time points to monitor the one or more sample quality metrics over time. In some embodiments, the method includes determining values of the one or more sample quality metrics substantially in real time.

[0057] In some embodiments, a machine learning model is used to calculate at least one specific sample quality metric of the one or more sample quality metrics, the machine learning model receiving one or more IR spectra as inputs and generating the specific sample quality metric as an output. In some embodiments, calculating the specific sample quality metric includes deconvolving an amide spectral region into sub-bands and / or calculating a second derivative spectrum.

[0058] In some embodiments, the IR absorbance data includes an IR absorbance spectrum, and step (c) includes: receiving (e.g., and / or accessing) one or more reference spectra, each reference spectrum measured from a corresponding (e.g., high-quality) reference sample [e.g., including a target viral vector species of high purity and / or concentration and / or one or more of its model components (e.g., model protein solution; e.g., model ssDNA solution)]; and using the IR absorbance spectrum and the one or more reference spectra (e.g., automatically) to determine at least a portion of the values of the one or more sample quality metrics [e.g., determining one or more (e.g., multiple) measures of the deviation between the reference spectrum and the IR absorbance spectrum as part of the values of the one or more viral vector sample quality metrics].

[0059] In some embodiments, the one or more reference spectra include high-quality viral vector spectra measured from a reference sample having a full capsid fraction equal to or higher than a specific threshold fraction (e.g., known a priori; e.g., determined to be). In some embodiments, the threshold fraction is about 75% [e.g., about 80% (e.g., about 90%)].

[0060] In certain embodiments, step (c) includes calculating a difference spectrum based on at least one of the one or more reference spectra and the IR absorbance spectrum (e.g., by subtracting the IR absorbance spectrum and / or its scaled or otherwise preprocessed version from the reference spectrum and / or its scaled or otherwise preprocessed version, or vice versa). In certain embodiments, step (c) includes calculating one or more derivative spectra of at least one of the one or more reference spectra and / or the IR absorbance spectrum (e.g., first derivative; e.g., second derivative). In certain embodiments, step (c) includes calculating (e.g., as one or more of the sample quality metrics) one or more members selected from the group consisting of: a correlation value based on the correlation between (i) a particular one of the one or more reference spectra and / or one or more of its derivatives and (ii) the IR absorbance spectrum and / or one or more of its derivatives; a covariance value based on the covariance between (i) a particular one of the one or more reference spectra and / or one or more of its derivatives and (ii) the IR absorbance spectrum and / or one or more of its derivatives; a Pearson's correlation value between (i) a particular one of the one or more reference spectra and / or one or more of its derivatives and (ii) the IR absorbance spectrum and / or one or more of its derivatives; and an overlap integral value based on the overlap integral between (i) a particular one of the one or more reference spectra and / or one or more of its derivatives and (ii) the IR absorbance spectrum and / or one or more of its derivatives.

[0061] In certain embodiments, step (c) includes: determining a value of a set of one or more specific peak metrics from the IR absorbance spectrum, thereby obtaining a set of sample peak metric values; and determining a value of one or more sample quality metrics based on the set of sample peak metric values and a set of reference peak metric values that have been determined for one or more specific peak metrics from one or more reference spectra.

[0062] In certain embodiments, the provided method includes determining (e.g., as one or more of the sample quality metric values) a similarity score that measures the similarity between one or more reference spectra and the IR absorbance spectrum. In certain embodiments, the method includes repeatedly performing steps (a)-(c) substantially in real time, thereby monitoring in real time for deviations from the one or more reference spectra.

[0063] In certain embodiments, the one or more time points are a plurality of time points [e.g., step (a) includes (e.g., repeatedly) measuring the IR absorption signal at each of the plurality of time points (e.g., continuously, in real time)].

[0064] In some embodiments, the provided method includes determining the value of a first sample quality metric at each of a plurality of time points and determining the value of a second (e.g., time difference; e.g., time aggregation) sample quality metric using the values of the first sample quality metric corresponding to two or more of the plurality of time points.

[0065] In some embodiments, the second sample quality metric is a time difference metric that measures the change in time of the first sample quality metric and is calculated based on the difference between (i) the value of the first sample quality metric for a first set of time points and (ii) the value of the first sample quality metric for a second set of time points [e.g., the difference between the value of the first sample quality metric at a first (e.g., current) time point and the value of the first sample quality metric at a second (e.g., previous) time point (e.g., the difference between the values of consecutive time points)].

[0066] In some embodiments, the second sample quality metric is a (e.g., real-time) time aggregation signal that is a function (e.g., a running sum, average, median, mode, variance, standard deviation, etc. over a particular time window) of at least a portion of the plurality of time points [e.g., a cumulative increasing portion, e.g., starting at a particular time point and ending at the current time point; e.g., a time window of a particular size (e.g., a look-back window)].

[0067] In some embodiments, step (c) includes causing, by a processor, the transmission of one or more trigger signals (e.g., voltages) to a controller unit of a production unit. In some embodiments, the one or more trigger signals include an analog voltage signal having a time-varying amplitude based on at least a portion of the value of one or more sample quality metrics (e.g., substantially proportional thereto).

[0068] In some embodiments, step (c) includes using a machine learning model to adjust one or more process parameters [e.g., wherein the machine learning model receives one or more sample quality metrics as inputs and generates an adjustment to a target process parameter and / or the target process parameter as an output; e.g., wherein the machine learning model receives one or more IR spectra as inputs and generates an adjustment to a target process parameter and / or the target process parameter as an output].

[0069] In some embodiments, the one or more process parameters include one or more members selected from the group consisting of flow rate, flow direction, pressure, temperature, and pH. In some embodiments, the one or more process parameters include the amount (e.g., absolute amount and / or relative amount) of one or more raw materials (e.g., used as inputs to a production unit). In some embodiments, the one or more process parameters include the time to initiate and / or stop a subprocess (e.g., heating, collecting elution fractions, growing, etc.).

[0070] In certain embodiments, when an aqueous sample exits a purification unit (e.g., a chromatography column), one or more IR absorbance signals corresponding to the IR absorbance data received in step (b) are measured from it (the aqueous sample) at each of one or more time points; and the method includes: using the IR absorption data to determine one or both of (i) the total capsid content and (ii) the full capsid fraction; and using the determined total capsid content and / or full capsid fraction to control the collection of the target fraction of the aqueous sample (e.g., during a specific collection window) to obtain a purified sample of the viral vector material.

[0071] In certain embodiments, when an aqueous sample exits a purification unit (e.g., a chromatography column), one or more IR absorbance signals corresponding to the IR absorbance data received in step (b) are measured from it (the aqueous sample) at each of one or more time points; and the method includes: using the IR absorption data to determine a capsid aggregation metric that measures the level of aggregation between capsids in a viral vector sample; and using the capsid aggregation metric to control the collection of the target fraction of the aqueous sample (e.g., during a specific collection window) to obtain a purified sample of the viral vector material.

[0072] In another aspect, the present disclosure relates to a system for obtaining a purified sample of a target protein species by real-time monitoring of protein heterogeneity and (e.g., automated; e.g., semi-automated) control of a purification process, the system including: (a) one or more mid-infrared (MIR) analyzers that are aligned and operable to measure, at each of one or more time points, the corresponding infrared (IR) absorbance signal of an aqueous sample exiting a purification unit (e.g., a chromatography column), the aqueous sample including one or more protein species that include the target protein species; (b) a processor of a computing device; and (c) a memory having instructions stored thereon, wherein the instructions, when executed by the processor, cause the processor to: receive IR absorbance data corresponding to the IR absorbance signal at each of the one or more time points; determine values of one or more sample quality metrics based on the IR absorbance data, the one or more sample quality metrics including a protein aggregation metric indicative of the level of protein aggregation within the aqueous sample; and provide (e.g., transmit to a controller unit of the purification unit) and / or use the one or more sample quality metrics to control the collection of the target fraction of the aqueous sample (e.g., during a specific collection window) to obtain the purified sample of the target protein species.

[0073] In certain embodiments, the provided system further includes a purification unit and / or its controller unit.

[0074] In another aspect, the present disclosure relates to a system for real-time monitoring of protein aggregation in a sample, the system comprising: a processor of a computing device; and a memory having instructions stored thereon, wherein the instructions, when executed by the processor, cause the processor to: (a) repeatedly receive IR absorbance data corresponding to IR absorbance signals measured at each of a plurality of time points, for each particular time point of the plurality of time points, the IR absorbance data including a corresponding IR absorbance spectrum measured from the sample at the particular time point and including a plurality of absorbance values, each absorbance value being associated with a particular wave number; (b) (e.g., automatically) analyze the IR absorbance data to obtain a real-time protein aggregation signal that provides a measure of protein aggregation in the sample over time, for each particular time point of the plurality of time points: determine values of one or more peak metrics for one or both of the amide I band and the amide II band using the IR absorbance spectrum corresponding to the particular time point; determine a value of a protein aggregation metric indicative of the level of protein aggregation in the sample at the particular time point using the values of the one or more peak metrics; and update the real-time protein aggregation signal with the determined value of the protein aggregation metric at the particular time point; and (c) store and / or provide the real-time protein aggregation signal for one or more of the following uses: (i) further processing, (ii) display, and (iii) use as a control signal for adjusting one or more purification units (e.g., a chromatography system).

[0075] In another aspect, the present disclosure relates to a system for monitoring and controlling a production unit for manufacturing a biological product (e.g., a protein; e.g., a virus) based on mid-infrared (MIR) spectroscopy, the method comprising: (a) one or more (e.g., integrated) mid-infrared (MIR) analyzers [e.g., the MIR analyzers described in one or more aspects and / or embodiments herein (e.g., in the above paragraphs)] that are aligned and operable to measure corresponding infrared (IR) absorbance signals of an aqueous sample flowing into and / or out of the production unit (e.g., and which includes one or more inputs, output products, waste products, or in-process products of the production unit) at each of one or more time points; (b) a processor of a computing device; and (c) a memory having instructions stored thereon, wherein the instructions, when executed by the processor, cause the processor to: receive IR absorbance data corresponding to the IR absorbance signals measured at each of the one or more time points; and use the received IR absorbance data to effect an adjustment of one or more process parameters of the production unit.

[0076] In certain embodiments, the provided system further includes a production unit and / or its controller unit.

[0077] In another aspect, the present disclosure relates to methods for (e.g., real-time) quantifying and / or monitoring the quality of viral vectors in an aqueous sample comprising one or more viruses and / or virus-like particles, the methods comprising: (a) receiving, by a processor of a computing device (e.g., repeatedly), IR absorbance data corresponding to one or more infrared (IR) absorbance signals measured from the sample; (b) determining, by the processor and using the IR absorbance data (e.g., automatically), values of one or more viral vector sample quality metrics; and (c) storing and / or providing the values of the one or more viral vector sample quality metrics for display and / or further processing.

[0078] In certain embodiments, one or more viral vector quality metrics include total capsid content that quantifies the capsid content within the sample [e.g., concentration (e.g., titer), mass per volume (e.g., mg / mL), number of (viral) particles per volume, etc.]. In certain embodiments, one or more viral vector sample quality metrics include full capsid fraction (e.g., percentage, ratio, etc. of full viral vectors). In certain embodiments, one or more viral vector sample quality metrics include a capsid aggregation metric that indicates the level of capsid aggregation within the viral vector sample. In certain embodiments, one or more viral vector sample quality metrics include a protein content metric that quantifies the protein content within the sample [e.g., concentration (e.g., titer), mass per volume (e.g., mg / mL), number of particles per volume, etc.].

[0079] In certain embodiments, one or more viral vector sample quality metrics include nucleic acid (e.g., DNA, RNA, etc.) content metrics that quantify the nucleic acid content within the sample [e.g., concentration (e.g., titer), mass per volume (e.g., mg / mL), number of particles per volume, viral genome copy number, etc.].

[0080] In certain embodiments, one or more viral vector sample quality metrics include viral nucleic acid (e.g., viral DNA, RNA, etc.) content metrics that distinguish viral nucleic acid from host cell protein and host cell nucleic acid content.

[0081] In certain embodiments, step (b) includes: determining, by a processor, the value of each of one or more peak metrics of the IR absorption data, where each peak metric is associated with one or more specific spectral bands [e.g., a continuous range of wavelengths / wavenumbers (e.g., amide-I band, amide-II band, amide-III band; e.g., amide region spanning two or more amide bands; e.g., asymmetric PO4 band; e.g., symmetric PO4 band)], and quantifying specific structural features of one or more absorption peaks within the specific spectral band [e.g., intensity (e.g., peak amplitude; e.g., area under the curve (AUC)); e.g., linewidth; e.g., frequency position (e.g., peak frequency; e.g., centroid frequency)]; and using the determined values of the one or more peak metrics to determine the value of at least a portion of the viral vector sample quality metric.

[0082] In certain embodiments, the IR absorbance data includes: (i) one or more (e.g., multiple) amide II absorbance values, which are associated with and measure the IR absorption of (a viral sample) at wavenumbers within the amide II spectral band (e.g., in the range of about 1500 cm -1 to about 1600 cm -1 , e.g., in the range of about 1500 cm -1 to about 1575 cm -1 ; e.g., in the range of about 1500 cm -1 to about 1550 cm -1 ; e.g., in the range of about 1540 cm -1 to about 1560 cm -1 ); and / or (ii) one or more (e.g., multiple) amide III absorbance values, which are associated with and measure the IR absorption of (a viral sample) at wavenumbers within the amide II spectral band (e.g., in the range of about 1250 cm -1 to about 1350 cm -1 , e.g., in the range of about 1250 cm -1 to about 1325 cm -1 ; e.g., in the range of about 1275 cm -1 to about 1325 cm -1 ; e.g., in the range of about 1280 cm -1 to about 1300 cm -1 ).

[0083] In certain embodiments, step (b) includes determining the value of a protein content metric [e.g., concentration (e.g., titer)] at least in part based on the amide II and / or amide III absorbance values, the protein content metric quantifying the protein content within the sample.

[0084] In certain embodiments, the provided method includes: determining a value of an amide II peak metric based on an amide II absorbance value and / or determining a value of an amide III peak metric based on an amide III absorbance value; and using the amide II peak metric value and / or the amide III peak metric value to determine a protein content metric value. In certain embodiments, the amide II peak metric and / or the amide III peak metric are peak intensity metrics that respectively quantify the intensity [e.g., peak height, area under the curve (AUC), etc.] of the amide II band and / or the amide III band.

[0085] In certain embodiments, one or more viral vector sample quality metrics include one or more protein structure metrics (e.g., protein structure metrics; e.g., protein tertiary and / or quaternary structure metrics) that indicate the presence and / or content (e.g., absolute content; e.g., relative content) of one or more specific protein structure forms (e.g., specific secondary structure motifs; e.g., specific tertiary and / or quaternary structure motifs / forms) within the sample (e.g., thereby providing monitoring of changes in the secondary / tertiary / quaternary structure of the capsid protein).

[0086] In certain embodiments, the IR absorbance data includes: (i) one or more (e.g., multiple) antisymmetric phosphate stretch (antisymmetric - PO4) absorbance values that are associated with and measure the IR absorption of the (viral sample) at a wavenumber within the antisymmetric - PO4 spectral band (e.g., in the range of about 1150 cm -1 to about 1250 cm -1 ; e.g., in the range of about 1175 cm -1 to about 1250 cm -1 ; e.g., in the range of about 1200 cm -1 to about 1250 cm -1 ; e.g., in the range of about 1210 cm -1 to about 1230 cm -1 ); and / or (ii) one or more (e.g., multiple) symmetric phosphate stretch (symmetric - PO4) absorbance values that are associated with and measure the IR absorption of the (viral sample) at a wavenumber within the symmetric - PO4 spectral band (e.g., in the range of about 1000 cm -1 to about 1100 cm -1 ; e.g., in the range of about 1050 cm -1 to about 1100 cm -1 ; e.g., in the range of about 1075 cm -1 to about 1100 cm -1 ; e.g., in the range of about 1075 cm -1 to about 1085 cm -1 ).

[0087] In certain embodiments, step (b) includes determining a value of a nucleic acid content metric (e.g., DNA, RNA, etc.) that quantifies the nucleic acid content [e.g., concentration (e.g., titer)] within the sample, based at least in part on the antisymmetric-PO4 and / or symmetric-PO4 absorbance values. In certain embodiments, the provided method includes determining a value of an antisymmetric-PO4 peak metric based on the antisymmetric-PO4 absorbance value and / or determining a value of a symmetric-PO4 peak metric based on the symmetric-PO4 absorbance value; and using the antisymmetric-PO4 peak metric value and / or the symmetric-PO4 peak metric value to determine the nucleic acid content metric value.

[0088] In certain embodiments, the antisymmetric-PO4 peak metric and / or the symmetric-PO4 peak metric are peak intensity metrics that respectively quantify the intensity [e.g., peak height, area under the curve (AUC), etc.] of the antisymmetric-PO4 band and / or the symmetric-PO4 band.

[0089] In certain embodiments, step (b) includes determining (i) a value of a protein content metric that quantifies the protein content [e.g., concentration (e.g., titer)] within the sample and (ii) a value of a nucleic acid (e.g., DNA, RNA, etc.) content metric that quantifies the nucleic acid content [e.g., concentration (e.g., titer)] within the sample, thereby independently quantifying the total protein and nucleic acid content within the sample. In certain embodiments, step (b) includes determining the total capsid content as one of the viral vector sample quality metrics, based at least in part on the value of the protein content metric (e.g., as a function thereof). In certain embodiments, step (b) includes determining the full capsid fraction as one of the viral vector sample quality metrics, based at least in part on (i) the value of the protein content metric and / or the total capsid content and (ii) the value of the nucleic acid content metric (e.g., as a function thereof).

[0090] In certain embodiments, the IR absorbance data is or includes one or more IR absorbance spectra, and for each specific wave number among a plurality of wave numbers spanning the measured spectral band, each IR absorbance spectrum includes a corresponding IR absorbance value that represents a measure of the absorption of IR light by the aqueous sample at the specific wave number. In certain embodiments, the measured spectral band spans one or more bands selected from the group consisting of the amide II band, the amide III band, the asymmetric-PO4 band, and the symmetric-PO4 band.

[0091] In certain embodiments, step (a) includes repeatedly receiving the IR absorbance data at a plurality of time points, thereby obtaining a corresponding set of IR absorbance data for each of the plurality of time points; and the provided method includes performing steps (b) through (c) on each set of IR absorbance data, thereby monitoring the total capsid content and / or the full capsid fraction over time.

[0092] In certain embodiments, the provided method includes performing steps (a) through (c) substantially in real time to obtain (i) a real-time capsid content signal and / or a full capsid fraction signal, the real-time capsid content signal providing a measure of the capsid content in a sample as a function of time, and the full capsid fraction signal providing a measure of the full capsid fraction in a sample as a function of time.

[0093] In certain embodiments, the provided method includes measuring one or more IR absorbance signals with one or more (e.g., integrated) mid-infrared (MIR) analyzers [e.g., the MIR analyzers described in various aspects and embodiments herein (e.g., in the above paragraphs)].

[0094] In certain embodiments, the provided method includes measuring a corresponding one of the one or more infrared (IR) absorbance signals at each of one or more time points.

[0095] In certain embodiments, the provided method includes measuring one or more IR absorbance signals of an aqueous sample as it flows (e.g., into) a production unit and / or out of a production unit (e.g., the aqueous sample includes one or more inputs, output products, waste products, or in-process products of the production unit).

[0096] In certain embodiments, the one or more viruses and / or virus-like particles include one or more adeno-associated viruses (AAVs). In certain embodiments, the one or more viruses include adenoviruses and / or retroviruses (e.g., lentiviruses). In certain embodiments, the one or more viruses include plant-based viruses (e.g., tobacco mosaic virus).

[0097] In certain embodiments, the one or more IR absorbance signals corresponding to the IR absorbance data are measured from an aqueous sample as it flows (e.g., into) a production unit and / or out of a production unit (e.g., the aqueous sample includes one or more inputs, output products, waste products, or in-process products of the production unit).

[0098] In certain embodiments, the production unit is a purification unit. In certain embodiments, the purification unit is a member of the group consisting of: an alternating tangential flow filtration (ATF) system, a tangential flow depth filtration (TFDF) system, a tangential flow filtration (TFF) system, a chromatography column, a direct or conventional flow filtration unit, an ultrafiltration unit, and a diafiltration unit. In certain embodiments, the purification unit is a chromatography column {e.g., and wherein the chromatography column is a member of the group consisting of: an affinity chromatography (AC) column, a hydrophobic interaction chromatography (HIC) column, an ion exchange chromatography (IEX) column, a size exclusion chromatography (SEC) column, and a mixed mode chromatography column [e.g., any combination of the foregoing chromatography columns (e.g., IEX and HIC; e.g., IEX and SEC)]}.

[0099] In certain embodiments, the production unit is or includes a bioreactor (e.g., a seed bioreactor; e.g., a production bioreactor).

[0100] In certain embodiments, step (c) includes causing, by a processor, generation of one or more trigger signals (e.g., voltages) and / or transmission of one or more trigger signals (e.g., voltages) to a controller unit of the production unit, at least in part based on one or more determined viral vector sample quality metrics (e.g., their values) [e.g., the value of the determined capsid content and / or the determined full capsid fraction (e.g., its value)]. In certain embodiments, step (c) includes causing, by a processor, generation of a trigger signal (e.g., an analog signal) having a value that is at least in part based on the determined viral vector quality metric [e.g., capsid content; e.g., full capsid fraction; e.g., capsid aggregation metric].

[0101] In certain embodiments, when the aqueous sample exits the purification unit (e.g., the chromatography column), one or more IR absorbance signals corresponding to the IR absorbance data received in step (a) are measured from it (the aqueous sample) at each of one or more time points; and the provided method includes: using the IR absorption data to determine one or both of (i) the total capsid content and (ii) the full capsid fraction; and using the determined total capsid content and / or full capsid fraction to control collection of the target fraction of the aqueous sample (e.g., during a specific collection window), thereby obtaining a purified sample of the viral vector material.

[0102] In certain embodiments, when the aqueous sample exits the purification unit (e.g., chromatography column), one or more IR absorbance signals corresponding to the IR absorbance data received in step (a) are measured from it (the aqueous sample) at each of one or more time points; and the provided method includes: using the IR absorption data to determine a capsid aggregation metric that measures the level of aggregation between capsids in the viral vector sample; and using the capsid aggregation metric to control the collection of the target fraction of the aqueous sample (e.g., during a specific collection window) to obtain a purified sample of the viral vector material.

[0103] In certain embodiments, the IR absorbance data includes an IR absorbance spectrum, and wherein step (b) includes: receiving (e.g., and / or accessing) one or more reference spectra, each reference spectrum measured from a corresponding (e.g., high-quality) reference sample comprising the target viral vector species and / or one or more of its model components at high purity and / or concentration (e.g., model protein solution; e.g., model ssDNA solution); and using the IR absorbance spectrum and the one or more reference spectra (e.g., automatically) to determine at least a portion of the value of one or more viral vector sample quality metrics [e.g., determining one or more (e.g., multiple) measures of the deviation between the reference spectrum and the IR absorbance spectrum as part of the value of one or more viral vector sample quality metrics].

[0104] In certain embodiments, the one or more reference spectra include a high-quality viral vector spectrum measured from a reference sample having (e.g., known a priori; e.g., determined to be) a full capsid fraction equal to or higher than a specific threshold fraction. In certain embodiments, the threshold fraction is about 75% [e.g., about 80% (e.g., about 90%)].

[0105] In certain embodiments, step (b) includes calculating a difference spectrum based on at least one of the one or more reference spectra and the IR absorbance spectrum (e.g., by subtracting the IR absorbance spectrum and / or its scaled or otherwise preprocessed version from the reference spectrum and / or its scaled or otherwise preprocessed version, or vice versa).

[0106] In certain embodiments, step (b) includes calculating one or more derivative spectra of at least one of the one or more reference spectra and / or the IR absorbance spectrum (e.g., first derivative; e.g., second derivative).

[0107] In some embodiments, step (b) includes calculating, e.g., as a measure of deviation, one or more members selected from the group consisting of: a correlation value based on the correlation of (i) a particular one of the one or more reference spectra and / or one or more of its derivatives and (ii) the IR absorbance spectrum and / or one or more of its derivatives; a covariance value based on the covariance of (i) a particular one of the one or more reference spectra and / or one or more of its derivatives and (ii) the IR absorbance spectrum and / or one or more of its derivatives; a Pearson correlation value between (i) a particular one of the one or more reference spectra and / or one or more of its derivatives and (ii) the IR absorbance spectrum and / or one or more of its derivatives; and an overlap integral value based on the overlap integral of (i) a particular one of the one or more reference spectra and / or one or more of its derivatives and (ii) the IR absorbance spectrum and / or one or more of its derivatives.

[0108] In some embodiments, step (b) includes: determining a value of a set of one or more specific peak metrics from the IR absorbance spectrum, thereby obtaining a set of sample peak metric values; and determining a measure of deviation based on the set of sample peak metric values and a set of reference peak metric values that have been determined for one or more specific peak metrics from one or more reference spectra.

[0109] In some embodiments, the provided method includes determining a similarity score as a measure of deviation, the similarity score measuring the similarity between one or more reference spectra and the IR absorbance spectrum.

[0110] In some embodiments, the provided method includes repeatedly performing steps (a)-(c) substantially in real time, whereby the deviation from one or more reference spectra is monitored in real time.

[0111] In another aspect, the present disclosure relates to certain methods for (e.g., real-time) evaluating and / or monitoring the quality of viral vector content within an aqueous sample comprising a target viral vector species, the methods comprising: (a) receiving, by a processor of a computing device (e.g., repeatedly), IR absorbance data corresponding to one or more infrared (IR) absorbance signals measured from the sample, the IR absorbance data comprising (e.g., at least one) IR absorbance spectrum measured from the sample and comprising a plurality of absorbance values, each absorbance value being associated with a specific wave number; (b) receiving, by the processor (e.g., and / or accessing) one or more reference spectra, each reference spectrum being measured from a corresponding (e.g., high-quality) reference sample comprising a target viral vector species of high purity and / or concentration and / or one or more of its model components (e.g., model protein solution; e.g., model ssDNA solution); (c) determining, by the processor, using the IR absorbance spectrum and the one or more reference spectra (e.g., automatically), one or more (e.g., a plurality of) measures of deviation; and (d) storing and / or providing the measure of deviation for display and / or further processing.

[0112] In another aspect, the present disclosure relates to a system for (e.g., real-time) quantifying and / or monitoring the quality of viral vectors within an aqueous sample comprising one or more viruses and / or virus-like particles, the system comprising: a processor of a computing device; and a memory having instructions stored thereon, wherein the instructions, when executed by the processor, cause the processor to perform the provided methods described in certain aspects and embodiments herein (e.g., in the paragraphs above).

[0113] In another aspect, the present disclosure relates to a method for monitoring compositional changes of a sample by infrared (IR) absorption spectroscopy (e.g., in real time), the method comprising: (a) repeatedly receiving, by a processor of a computing device, IR absorbance data corresponding to IR absorbance signals measured at each of a plurality of time points, wherein for each particular time point of the plurality of time points, the IR absorbance data includes a corresponding IR absorbance spectrum measured from the sample at the particular time point and includes a plurality of absorbance values, each absorbance value being associated with a particular wave number; (b) analyzing, by the processor (e.g., automatically), the IR absorbance data to obtain (e.g., in real time) a normalized spectral difference signal that measures the change in normalized spectral absorbance between consecutive time points, for each particular time point of the plurality of time points: normalizing the current IR absorbance spectrum corresponding to the particular time point using a reference absorbance value determined from values of the current IR absorbance spectrum at one or more reference wave numbers to obtain a current normalized spectrum; determining a current value of a spectral difference metric based on (e.g., calculated as) the difference between the current normalized spectrum and a previous normalized spectrum, wherein the previous normalized spectrum is based on one or more previously obtained IR absorption spectra (e.g., is a particular previously obtained IR absorption spectrum; e.g., is an average of a plurality of previously obtained IR absorption spectra), each previously obtained IR absorption spectrum corresponding to a particular previous time point (e.g., a particular time interval and / or multiples thereof before the current particular time point) and having been measured at the particular previous time point, and each particular previously obtained IR absorption spectrum having been normalized using a reference value determined from values of the particular previously obtained IR absorbance spectrum at the one or more reference wave numbers; and updating the real-time normalized spectral difference signal according to the current value of the normalized spectral difference metric; and (c) storing and / or providing, by the processor, the real-time normalized spectral difference signal for one or more of the following uses: (i) further processing, (ii) display, and (iii) use as a control signal for adjusting one or more process parameters of a production unit (e.g., a chromatography unit; e.g., a filtration unit).

[0114] In certain embodiments, the provided method includes (e.g., by the processor) identifying compositional changes of the sample (e.g., at a particular time) based on the real-time normalized spectral difference signal.

[0115] In certain embodiments, the provided method includes detecting a change point in the real-time normalized spectral difference signal [e.g., a change in statistical characteristics (e.g., mean, median, mode, variance, etc.); e.g., a step change], and identifying compositional changes of the sample based on the detected change point.

[0116] In certain embodiments, the provided method includes determining (e.g., at each time point, e.g., in real time) the value of one or more statistical parameters of a real-time normalized spectral difference signal (e.g., as a sample quality metric).

[0117] In certain embodiments, the one or more statistical parameters include one or more members selected from the group consisting of: mean [e.g., a running (e.g., backward-looking) mean, e.g., calculated as the mean of the real-time normalized spectral difference signal over a time window including (e.g., ending at) the current time point and one or more previous time points]; variance [e.g., a running (e.g., backward-looking) variance, e.g., calculated as the variance of the real-time normalized spectral difference signal over a time window including (e.g., ending at) the current time point and one or more previous time points]; mode; and standard deviation.

[0118] In certain embodiments, the provided method includes identifying a compositional change based on the value of at least one of the one or more statistical parameters (i) exceeding one or more thresholds and / or (ii) varying outside of a particular range [e.g., a predetermined threshold and / or range; e.g., during operation, e.g., during an initial stage of a process run (e.g., during an initial time window of a chromatographic run, e.g., during an initial ramp of a salt gradient, e.g., prior to protein elution) determined thresholds and / or ranges].

[0119] In certain embodiments, the sample is or includes an aqueous sample. In certain embodiments, the sample includes one or more protein species [e.g., a target protein species, such as a monoclonal antibody; e.g., as described in certain embodiments herein].

[0120] In certain embodiments, the provided method includes identifying (e.g., by a processor) a compositional change of the sample (e.g., at a particular time) corresponding to a change in the purity of a target protein species [e.g., the presence of protein species other than the target; e.g., the presence of an undesired form (e.g., non-monomeric) of the target protein species] and / or properties (e.g., secondary structure composition).

[0121] In certain embodiments, the identified compositional change is or includes (e.g., indicates) a change in the level and / or presence of protein aggregation within the aqueous sample. In certain embodiments, the identified compositional change is or includes (e.g., indicates) one or more members selected from the group consisting of: a change in the content (e.g., relative content) of one or more specific protein species within the (e.g., aqueous) sample; a change in the level and / or type of molecular conjugation (e.g., glycan, small molecule drug, polyethylene glycol, etc.); and a change in the content (e.g., α-helix content, β-sheet content, turn content, disordered content) of one or more protein secondary structure motifs.

[0122] In certain embodiments, the sample comprises nucleic acids (e.g., DNA, RNA, mRNA, etc.). In certain embodiments, the provided method comprises (e.g., by a processor) identifying (e.g., in an aqueous sample) a compositional change of the sample (e.g., at a particular time) that corresponds to a change in the purity and / or properties of the nucleic acids within the sample (e.g., a change in the relative amounts of one or more particular types of nucleic acids (e.g., DNA, RNA, ssDNA, dsDNA)).

[0123] In certain embodiments, the aqueous sample comprises one or more viruses and / or virus-like particles [e.g., adeno-associated virus vectors (AAV); e.g., lentiviral vectors]. In certain embodiments, the provided method comprises (e.g., by a processor) identifying a compositional change of the sample (e.g., at a particular time) that corresponds to a change in the purity and / or properties of the viruses and / or virus-like particles within the sample.

[0124] In certain embodiments, the compositional change corresponds to (e.g., indicates) a change in the relative fraction (e.g., percentage, ratio, etc.) of empty and / or full virus vectors within the aqueous sample.

[0125] In certain embodiments, the compositional change corresponds to (e.g., indicates) the level of capsid aggregation within the sample.

[0126] In certain embodiments, the compositional change corresponds to (e.g., indicates) a change in the relative amounts between viral nucleic acids from host cell proteins and host cell nucleic acid content.

[0127] In certain embodiments, the provided method comprises causing (e.g., triggering) an adjustment of one or more process parameters of a production unit based on (e.g., triggered by) the identification of a compositional change of the sample (e.g., detecting a change point; e.g., based on the value of one or more statistical parameters of a real-time normalized spectral difference signal).

[0128] In certain embodiments, the production unit is or includes a purification unit.

[0129] In certain embodiments, the purification unit is or includes a chromatography column.

[0130] In certain embodiments, the production unit is or includes one or more members selected from the group consisting of: a flow controller, a valve controller (e.g., for adjusting buffer composition (e.g., by valve switching)), a temperature controller (e.g., for adjusting one or more temperature set points and / or (e.g., time) profiles).

[0131] In certain embodiments, the provided method includes triggering a response of a production unit (e.g., any parameter discussed herein) [e.g., where the production unit is a first production unit and the sample is associated with a second (e.g., upstream or downstream) production unit (e.g., is an input, output, or component processed by the second production unit)].

[0132] In certain embodiments, the sample is an aqueous sample and the method includes causing an adjustment of a collection window to control collection of a target fraction (e.g., monomeric species of a protein) of the aqueous sample, thereby obtaining a purified sample.

[0133] In certain embodiments, the production unit is or includes a filtration unit (e.g., an ultrafiltration and / or diafiltration unit) (e.g., and where the method includes causing an adjustment of flow rate, transmembrane pressure, processing time, etc. to control the composition of the retentate and / or permeate).

[0134] In certain embodiments, the method includes monitoring the progress of a chemical reaction based on a real-time normalized spectral difference signal (e.g., identifying compositional changes using the real-time normalized spectral difference signal) [e.g., within a production unit (e.g., a bioreactor, transfection unit, pegylation unit, antibody-drug conjugation unit)].

[0135] In certain embodiments, the method includes causing an adjustment of one or more members selected from the group consisting of on-line buffer preparation, mixing process (e.g., in a mixing tank), temperature controller.

[0136] In certain embodiments, the reference absorbance value is determined by the value of the current IR absorbance spectrum at a single reference wavenumber, and the previous reference value is determined by the value of the previous IR absorbance spectrum at a single reference wavenumber.

[0137] In certain embodiments, determining the current value of the spectral difference metric includes calculating the integrated absorbance over one or more specific spectral bands (e.g., and subtracting the integrated absorbance values) for each of the current and previous normalized spectra.

[0138] In certain embodiments, the one or more specific spectral bands include one or more members selected from the group consisting of the amide I spectral band (e.g., in the range of about 1600 cm -1 to about 1700 cm -1 or about 1800 cm -1 (e.g., in the range of about 1600 cm -1 to about 1725 cm -1 ; e.g., in the range of about 1625 cm -1 to about 1725 cm -1 ; e.g., in the range of about 1630 cm -1to about 1650 cm -1 ), and / or the amide II region, ranging from about 1500 to about 1600 cm -1 (e.g., ranging from about 1500 cm -1 to about 1575 cm -1 ; e.g., ranging from about 1500 cm -1 to about 1550 cm -1 ; e.g., ranging from about 1540 cm -1 to about 1560 cm -1 )); the amide II spectral band (e.g., ranging from about 1500 cm -1 to about 1600 cm -1 , e.g., ranging from about 1500 cm -1 to about 1575 cm -1 ; e.g., ranging from about 1500 cm -1 to about 1550 cm -1 ; e.g., ranging from about 1540 cm -1 to about 1560 cm -1 )); and the amide III spectral band (e.g., ranging from about 1250 cm -1 to about 1350 cm -1 , e.g., ranging from about 1250 cm -1 to about 1325 cm -1 ; e.g., ranging from about 1275 cm -1 to about 1325 cm -1 ; e.g., ranging from about 1280 cm -1 to about 1300 cm -1 ).

[0139] In certain embodiments, one or more specific spectral bands include one or more members selected from the group consisting of: the asymmetric - PO4 spectral band (e.g., ranging from about 1150 cm -1 to about 1250 cm -1 , e.g., ranging from about 1175 cm -1 to about 1250 cm -1 ; e.g., ranging from about 1200 cm -1 to about 1250 cm -1 ; e.g., ranging from about 1210 cm -1 to about 1230 cm -1 ); and the symmetric - PO4 spectral band (e.g., ranging from about 1000 cm -1 to about 1100 cm -1 , e.g., ranging from about 1050 cm -1 to about 1100 cm -1 ; e.g., ranging from about 1075 cm-1 to about 1100 cm -1 ; for example, the range is from about 1075 cm -1 to about 1085 cm -1 ).

[0140] In certain embodiments, measuring an IR absorbance signal includes using one or more MIR analyzers (e.g., as described in any one of claims 9 to 24).

[0141] In another aspect, the present disclosure provides a method for monitoring temporal changes of a sample by infrared (IR) absorption spectroscopy (e.g., in real time), the method comprising: (a) repeatedly receiving, by a processor of a computing device, IR absorbance data corresponding to IR absorbance signals measured at each of a plurality of time points, wherein for each particular time point of the plurality of time points, the IR absorbance data includes a corresponding IR absorbance spectrum measured from the sample at the particular time point and includes a plurality of absorbance values, each absorbance value being associated with a particular wave number; (b) analyzing, by the processor (e.g., automatically), the IR absorbance data to obtain one or both of the following: a (e.g., real-time) time-differential signal that measures a temporal change between values of one or more features (e.g., a sample quality metric) determined using (i) a first set of one or more IR absorbance spectra corresponding to a first set of particular time points and (ii) a second set of one or more IR absorbance spectra corresponding to the first set of particular time points; and a (e.g., real-time) time-aggregation signal that is a function of at least a portion of the plurality of time points [e.g., a cumulative increasing portion, e.g., starting at a particular time point and ending at the current time point; e.g., a time window of a particular size (e.g., a look-back window)] (e.g., a running sum, average, median, mode, variance, standard deviation, etc. over a particular time window); and (c) storing and / or providing, by the processor, the time-differential signal and / or the time-aggregation signal for one or more of the following uses: (i) further processing, (ii) display, and (iii) use as a control signal for adjusting one or more process parameters of a production unit (e.g., a chromatography unit; e.g., a filtration unit).

[0142] In another aspect, the present disclosure provides a system for monitoring temporal (e.g., compositional) changes of a sample by infrared (IR) absorption spectroscopy (e.g., in real time), the system comprising: a processor of a computing device; and a memory having instructions stored thereon, wherein the instructions, when executed by the processor, cause the processor to perform the various methods described herein (e.g., in the above paragraphs).

[0143] In certain embodiments, the provided system further includes one or more MIR analyzers (e.g., as described in the above paragraphs).

[0144] In some aspects, the present disclosure provides methods for obtaining a purified sample of a target protein species by bioprocess monitoring and control based on mid-infrared (IR) spectroscopy. The provided methods include: (a) at each of a plurality of time points, measuring, by one or more mid-infrared (MIR) analyzers, the corresponding mid-infrared absorbance spectrum of an aqueous sample exiting a purification unit, the aqueous sample comprising one or more protein species, the protein species comprising the target protein species, thereby measuring a plurality of mid-infrared absorbance spectra over time; (b) receiving, by a processor of a computing device, spectral data corresponding to the plurality of measured mid-infrared absorbance spectra; (c) for each of at least a portion of the plurality of time points, determining, by the processor, a corresponding value of one or more sample quality metrics based on the spectral data, the one or more sample quality metrics including a measure of the concentration and / or purity of the target protein species in the aqueous sample; and (d) using the determined values of the one or more sample quality metrics to control the collection of a target fraction of the aqueous sample (e.g., during a specific collection window), thereby obtaining a purified sample of the target protein species.

[0145] In certain embodiments, the purification unit is or includes a chromatography column {e.g., and wherein the chromatography column is a member of the group consisting of: an affinity chromatography (AC) column, a hydrophobic interaction chromatography (HIC) column, an ion exchange chromatography (IEX) column, a size exclusion chromatography (SEC) column, and a mixed-mode chromatography column [e.g., any combination of the foregoing chromatography columns (e.g., IEX and HIC; e.g., IEX and SEC)]}. In certain embodiments, the purification unit is or includes an ultrafiltration and diafiltration system (UF / DF) [e.g., a tangential flow filtration (TFF) system].

[0146] In certain embodiments, the target protein species is selected from the group consisting of: monoclonal antibodies (mAbs), fusion proteins, viral capsid proteins, antibody-drug conjugates, recombinant proteins, and plasma proteins.

[0147] In certain embodiments, the aqueous sample includes multiple different molecular forms of a specific protein [e.g., a therapeutic protein (e.g., an mAb)], including monomeric forms and one or more aggregated forms (e.g., dimers and / or other multimers), and wherein the target protein species is the monomeric form of the specific protein.

[0148] In certain embodiments, the aqueous sample includes one or more subspecies of a specific protein, each subspecies having a specific desired level and / or type of molecular conjugation (e.g., glycans, small molecule drugs, polyethylene glycol, etc.), and wherein the target protein species is a specific one of the one or more subspecies.

[0149] In certain embodiments, one or more MIR analyzers include a quantum cascade laser (QCL)-based mid-infrared spectrometer, the QCL-based mid-infrared spectrometer including: a QCL-based source that is aligned and operable to emit an MIR beam [e.g., including one or more wavelengths substantially within the MIR spectral range (e.g., in the range of about 5000 cm -1 to about 500 cm -1 (e.g., about 2 to 20 microns))]; one or more sampling optics that are aligned to direct and / or allow the MIR beam and / or at least a portion thereof to pass through and / or contact at least a portion of an aqueous sample [e.g., wherein the MIR beam contacts the portion of the aqueous sample by reflection at an interface between a solid material (e.g., an ATR crystal and / or an optical fiber) and the aqueous sample (e.g., wherein the MIR beam undergoes total internal reflection and contacts / probes the portion of the aqueous sample through an evanescent wave extending into the aqueous sample)], and after passing through or contacting the portion of the aqueous sample, toward one or more detectors; and the one or more detectors that are aligned and operable to detect the MIR beam after the MIR beam has passed through and / or contacted the aqueous sample.

[0150] In certain embodiments, one or more sampling optics include a flow cell that includes a detection channel through which an aqueous sample flows; and one or more detectors that are aligned and operable to detect the MIR beam exiting the detection channel after the MIR beam has transmitted through the detection channel.

[0151] In certain embodiments, the QCL-based source is a tunable QCL that is operable to scan the emission frequency of the MIR beam through a plurality of frequencies within a scan range (e.g., wherein the scan range includes a range of about 1700 cm -1 to 1400 cm -1 ; e.g., wherein the scan range includes a range of 1300 cm -1 to 1050 cm -1 ; e.g., wherein the scan range includes at least 1200 cm -1 to 1000 cm -1within a range), and the method includes, at each of one or more time points: scanning the emission frequency of the MIR beam within the scan range of the tunable laser so as to irradiate the aqueous sample at a plurality of emission frequencies; and detecting the MIR beam at each of the plurality of emission frequencies with one or more detectors (e.g., the MIR beam that has (i) been internally reflected by the interface between the high refractive index material and the aqueous sample and / or (ii) transmitted through the detection channel through which the aqueous sample flows), thereby measuring a corresponding infrared (IR) spectrum including a plurality of values as the corresponding IR absorbance signal of the aqueous sample, each value among the plurality of values being associated with and representing and / or based on the power detected at a specific one of the plurality of emission frequencies.

[0152] In certain embodiments, the MIR analyzer is an in-line sensor (e.g., as opposed to an off-line or at-line sensor), and wherein step (a) includes repeatedly measuring the IR absorbance spectrum over time {e.g., every 20 seconds or less [e.g., every 10 seconds or less (e.g., every 5 seconds or less; (e.g., every second or less))]} as the aqueous solution exits the purification unit (e.g., thereby measuring the IR absorbance spectrum of the aqueous sample substantially in real time).

[0153] In certain embodiments, for each of one or more time points, the spectral data includes a corresponding amide band spectrum [e.g., for each specific wavelength among a plurality of sampling (e.g., emission) wavelengths in the range of about 1800 to about 800 cm -1 in the range (e.g., in the range of about 1700 to 1400 cm -1 in the range), the amide band spectrum includes a relevant absorption value representing the absorption level of a portion of the aqueous sample at the specific wavelength].

[0154] In certain embodiments, step (c) includes: receiving (e.g., and / or accessing) by a processor a reference spectrum of a target protein species, the reference spectrum having been measured from a specific corresponding reference sample including the target protein species that is substantially separated and / or of high purity [e.g., 75% purity or better (e.g., 90% purity or better (e.g., 95% purity or better))]; and repeatedly using the reference spectrum at each of the plurality of time points to determine the concentration of the target protein species within the aqueous sample at each time point, thereby tracking the concentration of the target protein species over time.

[0155] In certain embodiments, step (c) includes: receiving (e.g., and / or accessing) by a processor one or more impurity reference spectra, each impurity reference spectrum being associated with a particular impurity of interest and having been measured from a particular corresponding reference sample comprising the impurity of interest that is substantially separated and / or of high purity [e.g., 75% purity or better (e.g., 90% purity or better (e.g., 95% purity or better))]; and repeatedly using the one or more impurity reference spectra at each of a plurality of time points to determine the concentration of each of the impurities of interest within the aqueous sample.

[0156] In certain embodiments, for each of the one or more time points, the spectral data includes corresponding amide band spectra, and wherein step (c) includes determining the absorbance ratio at at least two wavenumbers within the amide band spectra as a measure of sample purity.

[0157] In certain embodiments, step (d) includes transmitting, by the processor, one or more trigger signals (e.g., voltages) to a controller unit of the purification unit and / or a downstream (downstream of the purification unit) valve.

[0158] In certain embodiments, step (d) includes one or both of the following: initiating, by the controller unit, collection of a target fraction of the aqueous sample based on one or more trigger signals [e.g., wherein a particular one of the one or more trigger signals is an analog signal, and the controller unit initiates collection of the target fraction based on the amplitude of the analog signal (e.g., when it exceeds a particular threshold; e.g., when it drops below a particular threshold); e.g., wherein a particular one of the one or more trigger signals is a digital signal that triggers (e.g., by transitioning from a 0 voltage level to a 1 voltage level, or vice versa) initiation of collection of the target fraction]; and stopping, by the controller unit, collection of the target fraction of the aqueous sample based on the one or more trigger signals [e.g., wherein a particular one of the one or more trigger signals is an analog signal, and the controller unit stops collection of the target fraction based on the amplitude of the analog signal (e.g., when it exceeds a particular threshold; e.g., when it drops below a particular threshold); e.g., wherein a particular one of the one or more trigger signals is a digital signal that triggers (e.g., by transitioning from a 0 voltage level to a 1 voltage level, or vice versa) stopping of collection of the target fraction].

[0159] In certain embodiments, the target protein species is the monomeric form of a specific protein (e.g., a monoclonal antibody), and the method includes: in step (c), determining over time the value of: (i) the concentration of the monomeric form of the specific protein and / or (ii) the cumulative purity of the monomeric form of the specific protein within the total collected volume of the sample exiting the purification unit [e.g., the relative fraction (e.g., mass) of the monomeric form of the specific protein collected relative to the total protein collected]; and in step (d), stopping the collection of the aqueous sample exiting the purification unit at a specific stopping time, at least in part based on the value of the concentration and / or cumulative purity of the monomeric form of the specific protein.

[0160] In certain embodiments, the aqueous sample includes (i) one or more highly aggregated forms of the specific protein and / or (ii) one or more fragmented species of the specific protein, and wherein step (c) includes determining over time the concentration of one or more aggregated forms of the specific protein and / or the concentration of one or more fragmented species.

[0161] In certain embodiments, the aqueous sample includes one or more excipients, and the method includes: in step (c), determining based on spectral data the value of the concentration and / or amount of one or more excipients within the aqueous sample exiting the purification unit at one or more time points; and in step (d), using the determined value of the excipient concentration and / or amount to control the collection of the target fraction of the aqueous sample.

[0162] In some aspects, the present disclosure provides methods for preparing biopharmaceutical formulations comprising one or more excipients, the provided methods comprising: (a) receiving a solution comprising a purified active pharmaceutical ingredient, the purified active pharmaceutical ingredient comprising a protein species (e.g., a monoclonal antibody); (b) injecting and / or mixing one or more excipients into the solution of the purified active pharmaceutical ingredient over a period of time to produce an in-process active pharmaceutical ingredient solution comprising the purified active pharmaceutical ingredient and the one or more excipients, wherein when the one or more excipients are injected and / or mixed, the relative concentrations of the purified active pharmaceutical ingredient and the one or more excipients vary over a period of time; (c) measuring, at each of one or more time points, one or both of the following by one or more mid-infrared (MIR) analyzers: (i) a corresponding mid-infrared absorbance spectrum from the in-process active pharmaceutical ingredient solution; and (ii) a corresponding mid-infrared absorbance spectrum from a stock solution comprising at least one of the one or more excipients, thereby measuring one or more mid-infrared absorbance spectra; (d) receiving, by a processor of a computing device, spectral data corresponding to the one or more measured mid-infrared absorbance spectra; (e) for each of at least a portion of the one or more time points, determining, by the processor, corresponding values of one or more sample quality metrics based on the spectral data, the one or more sample quality metrics comprising one or more measures of (i) the concentration and / or purity of the protein species and / or (ii) a subset of the one or more excipients; and (f) using the determined values of the one or more sample quality metrics to control the injection and / or mixing of the one or more excipients to obtain a final active pharmaceutical ingredient having a desired protein and / or excipient content and / or purity.

[0163] In certain embodiments, step (b) comprises using an ultrafiltration / diafiltration (UF / DF) system (e.g., for buffer exchange).

[0164] In certain embodiments, the protein species is or comprises a monoclonal antibody.

[0165] In certain embodiments, the one or more excipients are or comprise one or more surfactants {e.g., detergents; e.g., wetting agents and / or solubilizers [e.g., polysorbate 20 (Tween 20), polysorbate 80 (Tween 80), poloxamer (Pluronic F68 and F127), Triton X-100, Brij 30, Brij 35, etc.]}.

[0166] In certain embodiments, one or more excipients are or include one or more fillers {e.g., sugars and / or polyols [e.g., sucrose, trehalose, glucose, lactose, sorbitol, mannitol, glycerol, etc.]; e.g., amino acids [e.g., arginine, aspartic acid, glutamic acid, lysine, proline, glycine, histidine, methionine, alanine, etc.]; e.g., polymers and proteins [e.g., gelatin, PVP, PLGA, PEG, dextran, cyclodextrin and derivatives, starch derivatives, HSA, BSA]}.

[0167] In certain embodiments, one or more MIR analyzers include a quantum cascade laser (QCL)-based mid-infrared spectrometer, the QCL-based mid-infrared spectrometer including: a QCL-based source that is aligned and operable to emit an MIR beam [e.g., including one or more wavelengths that are substantially within the MIR spectral range (e.g., in the range of about 5000 cm -1 to about 500 cm -1 (e.g., about 2 to 20 microns))]; one or more sampling optics that are aligned to direct and / or allow the MIR beam and / or at least a portion thereof to pass through and / or contact at least a portion of a stock solution and / or a portion of an in-process API solution [e.g., wherein the MIR beam contacts the at least a portion of the stock solution and / or the portion of the in-process API solution by reflection at an interface between a solid material (e.g., ATR crystal and / or optical fiber) and the at least a portion of the stock solution and / or the portion of the in-process API solution (e.g., wherein the MIR beam undergoes total internal reflection and contacts / detects the at least a portion of the stock solution and / or the portion of the in-process API solution through an evanescent wave extending into the aqueous sample)], and after passing through or contacting the at least a portion of the stock solution and / or the portion of the in-process API solution, toward one or more detectors; and the one or more detectors that are aligned and operable to detect the MIR beam after the MIR beam passes through and / or contacts the at least a portion of the stock solution and / or the portion of the in-process API solution.

[0168] In certain embodiments, one or more sampling optics include a flow cell that includes a detection channel through which a portion of a stock solution and / or a portion of an in-process API solution flows; and the one or more detectors are aligned and operable to detect the MIR beam exiting the detection channel after the MIR beam has transmitted through the detection channel.

[0169] In certain embodiments, the QCL-based source is a tunable QCL that is operable to scan the emission frequency of the MIR beam through a plurality of frequencies within a scan range (e.g., wherein the scan range includes about 1700 cm-1 in the range of 1400 cm -1 ; for example, wherein the scan range includes 1300 cm -1 to 1050 cm -1 ; for example, wherein the scan range includes at least 1200 cm -1 to 1000 cm -1 (the range), and the method includes, at each of one or more time points: scanning the emission frequency of the MIR beam within the scan range of the tunable laser, thereby irradiating a portion of the stock solution and / or a portion of the in-process API solution at a plurality of emission frequencies; and detecting the MIR beam at each of the plurality of emission frequencies with one or more detectors (e.g., the MIR beam that has (i) been internally reflected by an interface between a high refractive index material and the portion of the stock solution and / or the portion of the in-process API solution and / or (ii) passed through a detection channel through which the portion of the stock solution and / or the portion of the in-process API solution has flowed), thereby measuring a corresponding infrared (IR) spectrum including a plurality of values as the corresponding IR absorbance signal of the portion of the stock solution and / or the portion of the in-process API solution, each value being associated with and representing and / or based on the power detected at a particular one of the plurality of emission frequencies.

[0170] In certain embodiments, the MIR analyzer is an in-line sensor (e.g., as opposed to an off-line or at-line sensor), and wherein step (c) includes repeatedly measuring the IR absorbance spectrum over time {e.g., every 20 seconds or less [e.g., every 10 seconds or less (e.g., every 5 seconds or less; (e.g., every second or less))]} when one or more excipients are being injected and / or mixed (e.g., thereby measuring the IR absorbance spectrum of the in-process API solution substantially in real time).

[0171] In certain embodiments, for each of one or more time points, the spectral data includes a corresponding amide band spectrum [e.g., for each particular wavelength of a plurality of sampling (e.g., emission) wavelengths in the range of about 1800 to about 800 cm -1 (e.g., in the range of about 1700 to 1400 cm -1 ), the amide band spectrum includes a relevant absorption value representing the absorption level of a portion of the aqueous sample at the particular wavelength]. In certain embodiments, for each of one or more time points, the spectral data includes a corresponding sugar band spectrum [e.g., for each of a plurality of sampling (e.g., emission) wavelengths in the range of about 1400 to about 800 cm -1 (e.g., in the range of about 1200 to 1000 cm -1For each particular wavelength of a plurality of sampling (e.g., emission) wavelengths (within the range of), the sugar band spectrum includes a relevant absorption value representing the absorption level of a portion of the aqueous sample at the particular wavelength.

[0172] In certain embodiments, step (e) includes: receiving (e.g., and / or accessing) by a processor a reference spectrum of a protein species, the reference spectrum having been measured from a particular corresponding reference sample comprising the protein species in substantially isolated and / or high purity [e.g., 75% purity or better (e.g., 90% purity or better (e.g., 95% purity or better))]; and repeatedly using the reference spectrum at each of a plurality of time points to determine the concentration of the protein species within the drug substance solution during each time point, thereby tracking the concentration of the protein species over time.

[0173] In certain embodiments, step (e) includes: receiving (e.g., and / or accessing) by a processor one or more excipient reference spectra, each excipient reference spectrum being associated with a particular excipient of interest (among one or more excipients) and having been measured from a particular corresponding reference sample comprising the particular excipient of interest in substantially isolated and / or high purity [e.g., 75% purity or better (e.g., 90% purity or better (e.g., 95% purity or better))]; and repeatedly using the one or more excipient reference spectra at each of a plurality of time points to determine the concentration of each of the one or more excipients of interest within the stock solution and / or the drug substance solution during the process.

[0174] In certain embodiments, step (f) includes causing by a processor the transmission of one or more trigger signals (e.g., voltage) to a controller unit (e.g., the controller unit of a UF / DF system; e.g., the controller unit of one or more valves).

[0175] In some embodiments, step (f) includes one or both of the following: initiating the injection and / or mixing of one or more excipients by the controller unit [e.g., where a particular one of the one or more trigger signals is an analog signal and the controller unit initiates injection and / or mixing based on the amplitude of the analog signal (e.g., when it exceeds a particular threshold; e.g., when it drops below a particular threshold); e.g., where a particular one of the one or more trigger signals is a digital signal that triggers (e.g., by transitioning from a 0 voltage level to a 1 voltage level, or vice versa) the initiation of injection and / or mixing], and stopping the injection and / or mixing of the one or more excipients by the controller unit based on the one or more trigger signals [e.g., where a particular one of the one or more trigger signals is an analog signal and the controller unit stops the injection and / or mixing of the one or more excipients based on the amplitude of the analog signal (e.g., when it exceeds a particular threshold; e.g., when it drops below a particular threshold); e.g., where a particular one of the one or more trigger signals is a digital signal that triggers (e.g., by transitioning from a 0 voltage level to a 1 voltage level, or vice versa) the stopping of the injection and / or mixing of the one or more excipients].

[0176] In some aspects, the present disclosure provides a system for obtaining a purified sample of a target protein species by bio-process monitoring and control based on mid-infrared (IR) spectroscopy. The provided system includes: one or more mid-infrared (MIR) analyzers [e.g., each MIR analyzer is operable to (e.g., based on one or more signals / communicate with a processor) measure a corresponding mid-infrared absorbance spectrum of an aqueous sample exiting a purification unit at each of a plurality of time points, the aqueous sample including one or more protein species that include the target protein species, thereby measuring a plurality of mid-infrared absorbance spectra over time]; a processor of a computing device; and a memory storing instructions thereon, where the instructions, when executed by the processor, cause the processor to: (a) receive spectral data corresponding to a plurality of measured mid-infrared absorbance spectra, each mid-infrared absorbance spectrum having been measured by the one or more MIR analyzers at a corresponding one of the plurality of time points from an aqueous sample exiting the purification unit, the aqueous sample including one or more protein species that include the target protein species; (b) for each of at least a portion of the plurality of time points, determine a corresponding value of one or more sample quality metrics based on the spectral data, the one or more sample quality metrics including a measure of the concentration and / or purity of the target protein species in the aqueous sample; and (c) use the determined values of the one or more sample quality metrics to control the collection of the target fraction of the aqueous sample (e.g., during a particular collection window), thereby obtaining a purified sample of the target protein species.

[0177] In some embodiments, the present disclosure provides a system for preparing a biopharmaceutical formulation comprising one or more excipients. The provided system includes: one or more mid-infrared (MIR) analyzers; a processor of a computing device; and a memory storing instructions thereon, wherein the instructions, when executed by the processor, cause the processor to: (a) receive spectral data corresponding to one or more mid-infrared absorbance spectra that have been measured by the one or more MIR analyzers at each of one or more time points from one or both of the following: (i) an in-process API solution that includes purified API and one or more excipients injected into and / or mixed with it over time; and (ii) a stock solution that includes at least one of the one or more excipients; (b) for each of at least a portion of the one or more time points, determine corresponding values of one or more sample quality metrics based on the spectral data, the one or more sample quality metrics including (i) protein species and / or (ii) one or more measures of the concentration and / or purity of a subset of the one or more excipients; and (c) use the determined values of the one or more sample quality metrics to control the injection and / or mixing of the one or more excipients to obtain a final API having a desired protein and / or excipient content and / or purity.

[0178] Features of embodiments described with respect to one aspect of the present disclosure may be applied with respect to another aspect of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0179] The foregoing and other objects, aspects, features, and advantages of the present disclosure will become more apparent and better understood by reference to the following description in conjunction with the accompanying drawings, in which:

[0180] Figure 1 A is a graph showing the absorption spectrum of a biomaterial according to an illustrative embodiment.

[0181] Figure 1B is a schematic diagram showing certain vibration modes according to an illustrative embodiment.

[0182] Figure 2 is a graph and schematic diagram showing nucleic acid absorption in the mid-infrared according to an illustrative embodiment.

[0183] Figure 3 is a schematic diagram showing the relationship between protein amide band vibrations and secondary structure conformations according to an illustrative embodiment.

[0184] Figure 4 is a graph comparing mid-infrared absorption spectrometry with UV absorption according to an illustrative embodiment.

[0185] Figure 5Shows two graphs of UV absorption according to an illustrative embodiment.

[0186] Fig. 6A Is a schematic diagram of an FT-IR spectrometer according to an illustrative embodiment.

[0187] Figure 6B Is a schematic diagram of a tunable QCL-based spectrometer according to an illustrative embodiment.

[0188] Figure 6C Is a schematic diagram showing the operation of an FT-IR spectrometer and a tunable QCL spectrometer according to an illustrative embodiment.

[0189] Fig.6D Is a schematic diagram showing the operation of an FT-IR spectrometer and a tunable QCL spectrometer according to an illustrative embodiment.

[0190] Fig. 6E Is a schematic diagram showing the operation of an FT-IR spectrometer and a tunable QCL spectrometer according to an illustrative embodiment.

[0191] Fig. 6F Is a schematic diagram showing the operation of an FT-IR spectrometer and a tunable QCL spectrometer according to an illustrative embodiment.

[0192] Figure 7 Is a graph showing the spectral brightness of certain mid-infrared sources according to an illustrative embodiment.

[0193] Figure 8 Is a graph showing water absorption in the mid-infrared according to an illustrative embodiment.

[0194] Fig. 9A Is a schematic diagram of a tunable QCL-based spectrometer according to an illustrative embodiment.

[0195] Fig. 9B Is a schematic diagram showing the tuning ranges of two QCL-based spectrometers according to an illustrative embodiment.

[0196] Fig. 10A Is a schematic diagram showing certain steps and mathematical processing for obtaining an absorbance spectrum of an analyte present in a mobile phase according to one or more illustrative embodiments.

[0197] Fig. 10B Is a schematic diagram showing certain steps and mathematical processing for obtaining an absorbance spectrum of an analyte present in a mobile phase according to one or more illustrative embodiments.

[0198] Fig. 10C Is a schematic diagram showing certain steps and mathematical processing for obtaining an absorbance spectrum of an analyte present in a mobile phase according to one or more illustrative embodiments.

[0199] Fig. 10D is a schematic diagram showing certain steps and mathematical processing for obtaining an absorbance spectrum of an analyte present in a mobile phase according to one or more illustrative embodiments.

[0200] Fig. 10E is a schematic diagram showing certain steps and mathematical processing for obtaining an absorbance spectrum of an analyte present in a mobile phase according to one or more illustrative embodiments.

[0201] Fig.10F is a schematic diagram showing certain steps and mathematical processing for obtaining an absorbance spectrum of an analyte present in a mobile phase according to one or more illustrative embodiments.

[0202] Figure 10G is a schematic diagram showing certain steps and mathematical processing for obtaining an absorbance spectrum of an analyte present in a mobile phase according to one or more illustrative embodiments.

[0203] Fig. 10H is a schematic diagram showing certain steps and mathematical processing for obtaining an absorbance spectrum of an analyte present in a mobile phase according to one or more illustrative embodiments.

[0204] Fig.11 is an illustrative diagram of three IR absorbance spectra according to an illustrative embodiment, which shows the absorbance of a protein-nucleic acid mixture and the spectra of the individual protein and nucleic acid components.

[0205] Fig.12 is a graph of IR absorbance spectra measured from high-quality viral vector samples (e.g., having a high proportion of full capsids) and lower-quality viral vector samples (e.g., including a large proportion of empty capsids) according to an illustrative embodiment, shown together with the difference spectrum.

[0206] Fig.13 is a schematic diagram showing measurements performed at various stages and production units in a biomanufacturing process using an exemplary mid-infrared spectrometer according to an illustrative embodiment.

[0207] Fig.14 is a schematic diagram showing an empty adeno-associated virus (AAV) capsid and an AAV capsid loaded with a gene cassette according to an illustrative embodiment.

[0208] Fig.15 is a block flow diagram of an exemplary process for using IR absorption data measured from a viral vector sample to determine a sample quality metric and / or control a bioproduction process according to an illustrative embodiment.

[0209] Fig.16AIt is a diagram showing the steps in the manufacturing process of an AAV vector according to an illustrative embodiment.

[0210] Fig. 16B It is a diagram showing the steps in the manufacturing process of a lentiviral vector according to an illustrative embodiment.

[0211] Fig.17 It is a block flow diagram of an exemplary process for using a reference spectrum to determine a sample quality metric and / or control a biological production process according to an illustrative embodiment.

[0212] Fig.18 It is a block diagram of an exemplary cloud computing environment used in certain embodiments.

[0213] Fig.19 It is a block diagram of an exemplary computing device and an exemplary mobile computing device used in certain embodiments.

[0214] Fig. 20A It is a diagram showing the measurement of bovine serum albumin (BSA) using mid-infrared spectroscopy according to an illustrative embodiment.

[0215] Fig. 20B It is a diagram showing the measurement of bovine serum albumin (BSA) using mid-infrared spectroscopy according to an illustrative embodiment.

[0216] Fig. 20C It is a diagram showing the measurement of bovine serum albumin (BSA) using mid-infrared spectroscopy according to an illustrative embodiment.

[0217] Fig.20D It is a diagram showing the measurement of bovine serum albumin (BSA) using mid-infrared spectroscopy according to an illustrative embodiment.

[0218] Fig.21 It is a diagram showing the use of an absorption spectrum for chromatographic process monitoring according to an illustrative embodiment.

[0219] Fig.22A It is a diagram showing the use of an absorption spectrum for chromatographic process monitoring according to an illustrative embodiment.

[0220] Fig. 22B It is a diagram showing the use of an absorption spectrum for chromatographic process monitoring according to an illustrative embodiment.

[0221] Fig.23A It is a diagram showing the measurement of protein secondary structure content using mid-infrared spectroscopy according to an illustrative embodiment.

[0222] Fig. 23B It is a diagram showing the measurement of protein secondary structure content using mid-infrared spectroscopy according to an illustrative embodiment.

[0223] Fig.23C It is a figure showing the measurement of the content of protein secondary structure using mid-infrared spectroscopy according to an illustrative embodiment.

[0224] Fig.23D It is a figure showing the measurement of the content of protein secondary structure using mid-infrared spectroscopy according to an illustrative embodiment.

[0225] Fig.24 It is a set of figures showing the measurement of the content of protein secondary structure using mid-infrared spectroscopy according to an illustrative embodiment.

[0226] Fig.25 It shows the calculation of the relative density of a protein using mid-infrared absorption spectroscopy according to an illustrative embodiment.

[0227] Fig.26A It is a figure showing the change of the mid-infrared baseline with conductivity during elution according to an illustrative embodiment.

[0228] Fig.26B It is a figure showing the change of the mid-infrared baseline with conductivity during elution according to an illustrative embodiment.

[0229] Fig. 27 It is a figure showing the calculation of certain sample quality metrics described herein according to certain illustrative embodiments.

[0230] Fig.28A It is a figure showing the calculation of certain sample quality metrics described herein according to certain illustrative embodiments.

[0231] Fig.28B It is a figure showing the calculation of certain sample quality metrics described herein according to certain illustrative embodiments.

[0232] Fig.29A It is a figure showing the calculation of certain sample quality metrics described herein according to certain illustrative embodiments.

[0233] Fig.29B It is a figure showing the calculation of certain sample quality metrics described herein according to certain illustrative embodiments.

[0234] Fig. 30A It is a figure showing the calculation of certain sample quality metrics described herein according to certain illustrative embodiments.

[0235] Fig. 30B It is a figure showing the calculation of certain sample quality metrics described herein according to certain illustrative embodiments.

[0236] Fig.31 It is a figure showing the calculation of certain sample quality metrics described herein according to certain illustrative embodiments.

[0237] Fig.32A is a block flow diagram of a process for controlling a production process based on IR absorption spectroscopy according to an illustrative embodiment.

[0238] Fig.32B is an illustrative sketch showing the expected (hypothetical) variation over time of sample quality metrics measuring total protein content and protein aggregation during elution from an IEX column, as described herein, according to an illustrative embodiment.

[0239] Fig.33A is a diagram showing certain workflows and data processing methods for performing mid-infrared spectroscopy measurements according to various illustrative embodiments.

[0240] Fig.33B is a diagram showing certain workflows and data processing methods for performing mid-infrared spectroscopy measurements according to various illustrative embodiments.

[0241] Fig.33C is a diagram showing certain workflows and data processing methods for performing mid-infrared spectroscopy measurements according to various illustrative embodiments.

[0242] Fig.34A is a schematic diagram showing elution control according to an illustrative embodiment.

[0243] Fig.34B is a schematic diagram showing elution control according to an illustrative embodiment.

[0244] Fig.35 is a set of diagrams showing protein secondary structure monitoring during TFF injection according to an illustrative embodiment.

[0245] Fig.36 is a diagram showing the spectral coverage of a QCL system according to an illustrative embodiment.

[0246] Fig.37 is a diagram showing sensitivity improvement in a mid-infrared QCL system.

[0247] Fig.38A is a diagram showing the measurement of multiple analytes according to an illustrative embodiment.

[0248] Fig.38B is a diagram showing the measurement of multiple analytes according to an illustrative embodiment.

[0249] Fig.39A is a diagram showing the measurement of multiple analytes according to an illustrative embodiment.

[0250] Fig.39B is a diagram showing the measurement of multiple analytes according to an illustrative embodiment.

[0251] Fig.40A It is a diagram showing the measurement of multiple analytes according to an illustrative embodiment.

[0252] Fig.40B It is a diagram showing the measurement of multiple analytes according to an illustrative embodiment.

[0253] Fig.41A It is a diagram showing the measurement of protein secondary structure according to an illustrative embodiment.

[0254] Fig.41B It is a diagram showing the measurement of protein secondary structure according to an illustrative embodiment.

[0255] Fig.41C It is a diagram showing the measurement of protein secondary structure according to an illustrative embodiment.

[0256] Fig.42 It is a diagram showing the measurement of protein secondary structure according to an illustrative embodiment.

[0257] Fig.43A It is a diagram showing the measurement of protein secondary structure according to an illustrative embodiment.

[0258] Fig.43B It is a diagram showing the measurement of protein secondary structure according to an illustrative embodiment.

[0259] Fig.44 It is a schematic diagram of an exemplary system according to an illustrative embodiment including multiple mid-infrared analyzers and an external computer.

[0260] Fig.45A It is a diagram plotting the results of certain reproducibility tests according to an illustrative embodiment.

[0261] Fig.45B It is a diagram plotting the results of certain reproducibility tests according to an illustrative embodiment.

[0262] Fig.45C It is a diagram plotting the results of certain reproducibility tests according to an illustrative embodiment.

[0263] Fig.45D It is a diagram plotting the results of certain reproducibility tests according to an illustrative embodiment.

[0264] Fig.45E It is a diagram plotting the results of certain reproducibility tests according to an illustrative embodiment.

[0265] Fig.46A It is a diagram showing the IR absorbance spectra of DNA and protein (BSA) solutions according to an illustrative embodiment.

[0266] Fig.46BIt is a diagram showing the IR absorbance spectra of various DNA-protein mixtures according to an illustrative embodiment.

[0267] Fig.46C It is a diagram showing the molecular structure of nucleobases.

[0268] Fig.46D It is a graph of three IR absorbance spectra according to an illustrative embodiment, showing the absorbance of a protein-nucleic acid mixture and the spectra of the individual protein and nucleic acid components.

[0269] Fig.47A It is a diagram of IR absorbance spectra measured for samples containing various concentrations of monoclonal antibodies according to an illustrative embodiment.

[0270] Fig.47B It is a diagram of IR absorbance spectra measured for samples containing various concentrations of monoclonal antibodies according to an illustrative embodiment.

[0271] Fig.48 It is a diagram showing the absorbance-based chromatogram and the variation of the normalized spectral difference signal over time according to an illustrative embodiment.

[0272] Fig.49 It shows a set of diagrams according to an illustrative embodiment, showing the variation of the absorbance-based chromatogram and different versions of the normalized spectral difference signal over time.

[0273] Fig.50A It is a schematic diagram of a system for protein purification according to an illustrative embodiment.

[0274] Fig.50B It is a block flow diagram of a process for monitoring protein purification by mid-infrared spectroscopy according to an illustrative embodiment.

[0275] Fig.50C It is a block flow diagram of a process for determining component concentrations and / or sample quality metrics from measured IR spectra according to an illustrative embodiment.

[0276] Fig.51A It is a diagram of three reference spectra according to an illustrative embodiment.

[0277] Fig.51B It is a diagram of the simulated variation of the integrated absorbance over time during the operation of an ion exchange column according to an illustrative embodiment.

[0278] Fig.51C It is a diagram showing the protein components extracted from a simulated experiment according to an illustrative embodiment.

[0279] Fig.51DIt is a graph showing the changes in purity and yield determined from a simulated chromatography run according to an illustrative embodiment.

[0280] Fig.51E It is a set of three graphs according to an illustrative embodiment, showing (i) the extracted protein components, (ii) the integrated absorbance, and (iii) the purity and yield determined for a simulated chromatography run.

[0281] Fig.51F It is a set of three graphs according to an illustrative embodiment, showing (i) the extracted protein components, (ii) the integrated absorbance, and (iii) the purity and yield determined for a simulated chromatography run.

[0282] Figure 51G It is a set of three graphs according to an illustrative embodiment, showing (i) the extracted protein components, (ii) the integrated absorbance, and (iii) the purity and yield determined for a simulated chromatography run.

[0283] Fig.51H It is a set of three graphs according to an illustrative embodiment, showing (i) the extracted protein components, (ii) the integrated absorbance, and (iii) the purity and yield determined for a simulated chromatography run.

[0284] Fig.51I It is a graph showing the integrated absorbance measured during a size exclusion chromatography run according to an illustrative embodiment.

[0285] Fig.51J It is a graph showing the spectra of extracted dimers, monomers, and mixed dimer / monomer according to an illustrative embodiment.

[0286] Figure 51K It is a graph showing the fragment reference spectrum according to an illustrative embodiment.

[0287] Fig.52A It is a graph showing the mid-infrared spectra of certain buffers.

[0288] Fig.52B It is a graph showing the mid-infrared spectra of certain sugars.

[0289] Fig.53 It is a schematic diagram showing the monitoring of the UF / DF process by an on-line mid-infrared analyzer and an on-line UV spectrometer according to an illustrative embodiment.

[0290] Fig.54A It is a graph showing the BSA concentration measured over time determined by mid-infrared spectroscopy and UV absorbance.

[0291] Fig.54B It is a graph showing the change in sucrose concentration over time measured by mid-infrared spectroscopy and Cedex determination.

[0292] Fig.55A It is a diagram showing the mid-infrared spectra of various concentrations of polysorbate 80 (PS80) in water.

[0293] Fig.55B It is a diagram showing the mid-infrared spectra of PS80 incorporated into a formulation buffer at various concentrations.

[0294] Fig.56A It is a diagram showing the mid-infrared spectra of two sugars.

[0295] Fig.56B It is a diagram showing the mid-infrared spectra of two proteins within the sugar absorption band.

[0296] Fig.57A It is a diagram showing the mid-infrared spectra of the high-molecular-weight form of a monoclonal antibody incorporated into a pure monomer solution at different concentrations.

[0297] Fig.57B It is a diagram showing Fig.57A details around the amide band region of the spectra shown.

[0298] Fig.57C It is a diagram comparing the correlation of IR peak metrics with the monomer purity measured by size exclusion chromatography.

[0299] The features and advantages of the present disclosure will become more apparent from the following detailed description of certain embodiments, particularly when taken in conjunction with the accompanying drawings, in which like reference numerals always identify corresponding elements. In the drawings, like reference numerals generally indicate identical, functionally similar, and / or structurally similar elements. Detailed Description

[0300] Some definitions

[0301] To more easily understand the present disclosure, certain terms are first defined below. Additional definitions of the following terms and other terms are set forth throughout the specification.

[0302] A, an: As used herein, "a" or "an" with respect to a claim feature means "one or more" or "at least one".

[0303] Absorption data, absorption spectrum, absorbance data, absorbance spectrum: As used herein, the terms "absorption data", "absorption spectrum", "absorbance data", "absorbance spectrum" in, for example, "IR absorption data", "IR absorption spectrum", "IR absorbance data", "IR absorbance spectrum", etc. are used to refer to data, such as spectral data, that represents and / or indicates the optical absorption of a sample, regardless of whether the data and / or underlying signal is obtained by transmission measurement, attenuated total internal (ATR) measurement, reflection measurement, or other measurement. The use of the terms "absorption" and "absorbance" is not intended to be limited with respect to a particular sampling / measurement geometry, display and / or representation, or unit system. For example, an absorption spectrum can be represented as a transmission spectrum, where absorption bands appear as negative peaks, or as an absorption spectrum, where absorption peaks are positive and point upward. Absorption spectra can be represented in linear or logarithmic units. Although in some cases the term "absorbance" is used in IR spectroscopy to refer to a unit that is proportional to concentration and path length (e.g., the logarithm of transmittance), its use herein is not intended to limit any method, system, process, calculation, etc. to being performed in any particular unit system.

[0304] Administration: As used herein, the term "administration" generally refers to the administration of a composition to an individual or a system. Those of ordinary skill in the art will recognize the various routes that can be used to administer to an individual (e.g., a human) when appropriate. For example, in some embodiments, administration can be ocular, oral, parenteral, topical application, etc. In some specific embodiments, administration can be bronchial (e.g., by bronchial instillation), buccal, cutaneous (which can be or include, for example, one or more of dermal topical, intradermal, intercutaneous, transdermal, etc.), enteral, intraarterial, intradermal, intragastric, intramedullary, intramuscular, intranasal, intraperitoneal, intrathecal, intravenous, intraventricular, into a specific organ (e.g., intrahepatic), mucosal, nasal, oral, rectal, subcutaneous, sublingual, topical, tracheal (e.g., by tracheal instillation), vaginal, intravitreal administration, etc. In some embodiments, administration can involve intermittent dosing (e.g., multiple doses separated in time) and / or periodic dosing (e.g., individual doses separated by a common time period). In some embodiments, administration can involve continuous dosing (e.g., perfusion) for at least a selected period of time.

[0305] Affinity: As is known in the art, "affinity" is a measure of the tightness with which two or more binding partners associate with each other. Those skilled in the art are aware of the various assays that can be used to evaluate affinity and also know the appropriate controls for such assays. In some embodiments, affinity is evaluated in a quantitative assay. In some embodiments, affinity is evaluated at multiple concentrations (e.g., one binding partner at a time). In some embodiments, affinity is evaluated in the presence of one or more potential competing entities (e.g., competing entities that may be present in a relevant (e.g., physiological) environment). In some embodiments, affinity is evaluated relative to a reference (e.g., a reference ["positive control" reference] having a known affinity above a specific threshold or a reference ["negative control" reference] having a known affinity below a specific threshold). In some embodiments, affinity can be evaluated relative to a contemporaneous reference; in some embodiments, affinity can be evaluated relative to a historical reference. Generally, when evaluating affinity relative to a reference, the affinity is evaluated under comparable conditions.

[0306] Amino acid: In its broadest sense, as used herein, refers to any compound and / or substance that can be incorporated into a polypeptide chain, for example, by forming one or more peptide bonds. In some embodiments, an amino acid has the general formula structure H2N-C(H)(R)-COOH. In some embodiments, an amino acid is a naturally occurring amino acid. In some embodiments, an amino acid is a non-natural amino acid; in some embodiments, an amino acid is a D-amino acid; in some embodiments, an amino acid is an L-amino acid. A "standard amino acid" refers to any one of the twenty standard L-amino acids that are typically found in naturally occurring peptides. A "non-standard amino acid" refers to any amino acid other than a standard amino acid, whether prepared synthetically or obtained from natural sources. In some embodiments, compared to the above general formula structure, an amino acid (including the carboxyl and / or amino-terminal amino acids in a polypeptide) may contain structural modifications. For example, in some embodiments, compared to the general formula structure, an amino acid can be modified by methylation, amidation, acetylation, polyethylene glycolylation, glycosylation, phosphorylation, and / or substitution (e.g., substitution of an amino group, a carboxylic acid group, one or more protons, and / or a hydroxyl group). In some embodiments, compared to a polypeptide containing an otherwise identical unmodified amino acid, such a modification can, for example, alter the circulating half-life of the polypeptide containing the modified amino acid. In some embodiments, compared to a polypeptide containing an otherwise identical unmodified amino acid, such a modification does not significantly alter the relevant activity of the polypeptide containing the modified amino acid. As will be clear from the context, in some embodiments, the term "amino acid" can be used to refer to a free amino acid; in some embodiments, it can be used to refer to an amino acid residue of a polypeptide.

[0307] Antibodies, antibody polypeptides: As used herein, the terms "antibody polypeptide" or "antibody" or "antigen-binding fragment thereof" (which may be used interchangeably) refer to a polypeptide capable of binding to an epitope. In some embodiments, the antibody polypeptide is a full-length antibody, and in some embodiments, less than full-length but contains at least one binding site (including at least one, preferably at least two sequences having the structure of the "variable region" of an antibody). In some embodiments, the term "antibody polypeptide" encompasses any protein having a binding domain that is homologous or highly homologous to an immunoglobulin binding domain. In certain embodiments, an "antibody polypeptide" encompasses a polypeptide having a binding domain that exhibits at least 99% identity to an immunoglobulin binding domain. In some embodiments, an "antibody polypeptide" is any protein having a binding domain that exhibits at least 70%, 80%, 85%, 90% or 95% identity to an immunoglobulin binding domain (e.g., a reference immunoglobulin binding domain). The included "antibody polypeptides" may have the same amino acid sequence as the antibody found in its natural source. The antibody polypeptides according to the present invention can be prepared by any available means, including, for example, isolation from natural sources or antibody libraries, recombinant production in or with a host system, chemical synthesis, etc., or combinations thereof. The antibody polypeptides can be monoclonal or polyclonal. The antibody polypeptides can be members of any immunoglobulin class, including any human class: IgG, IgM, IgA, IgD, and IgE. In certain embodiments, the antibody can be a member of the IgG immunoglobulin class. As used herein, the terms "antibody polypeptide" or "characteristic part of an antibody" are used interchangeably and refer to any derivative of an antibody having the ability to bind to a target epitope. In certain embodiments, an "antibody polypeptide" is an antibody fragment that retains at least a substantial portion of the specific binding ability of the full-length antibody. Examples of antibody fragments include, but are not limited to, Fab, Fab', F(ab')2, scFv, Fv, dsFv, bispecific antibodies, and Fd fragments. Alternatively or additionally, the antibody fragment can include, for example, multiple chains linked together by disulfide bonds. In some embodiments, the antibody polypeptide can be a human antibody. In some embodiments, the antibody polypeptide can be humanized. A humanized antibody polypeptide can comprise a chimeric immunoglobulin, immunoglobulin chain, or antibody polypeptide (such as Fv, Fab, Fab', F(ab')2, or other antigen-binding sequences of an antibody) that contains a minimal sequence derived from a non-human immunoglobulin. Generally, a humanized antibody is a human immunoglobulin (recipient antibody) in which the residues of the complementarity-determining regions (CDRs) from the recipient are replaced with the residues of the CDRs from a non-human species (donor antibody) such as a mouse, rat, or rabbit having the desired specificity, affinity, and capacity.

[0308] About: As used herein, when applied to one or more values of interest, the term "about" or "approximately" refers to a value similar to the reference value. In certain embodiments, the term "about" or "approximately" refers to a range of values that fall within 25%, 20%, 19%, 18%, 17%, 16%, 15%, 14%, 13%, 12%, 11%, 10%, 9%, 8%, 7%, 6%, 5%, 4%, 3%, 2%, 1% or less in either direction (greater than or less than) of the reference value, unless otherwise stated or otherwise apparent from the context (unless such numbers exceed 100% of the possible value).

[0309] Backbone, peptide backbone: As used herein, the term "backbone", e.g., in a backbone or a peptide or polypeptide, refers to the portion of the peptide or polypeptide chain that includes the linkages between the amino acids of the chain but does not include the side chains. In other words, the backbone refers to the portion of the peptide or polypeptide that would remain upon removal of the side chains. In certain embodiments, the backbone is a chain in which the carboxyl group of one amino acid is bound to the amino group of the next amino acid by a peptide bond, and so on. The backbone may also be referred to as the "peptide backbone". It should be understood that in the case of using the term "peptide backbone", it is used for clarity and is not intended to limit the length of a particular backbone. That is, the term "peptide backbone" can be used to describe the peptide backbone of a peptide and / or a protein.

[0310] Biological product: As used herein, the term "biological product" refers to a composition that is produced by or can be produced by recombinant DNA technology, chemical synthesis, peptide synthesis, or purified and / or isolated from natural sources (such as humans, animals, or microorganisms) and has the desired biological activity. Biological products can be, for example, proteins, peptides, glycoproteins, polysaccharides, nucleic acids, phospholipids, mixtures of proteins or peptides, mixtures of glycoproteins, mixtures of polysaccharides, mixtures of nucleic acids, mixtures of one or more of proteins, peptides, glycoproteins, polysaccharides, or nucleic acids, or derivative forms and / or assemblies of any of the foregoing entities. In certain embodiments, biological products can be or include living entities, such as cells or tissues. The molecular weight of biological products can vary widely, from about 1000 Da for small peptides such as peptide hormones to 1000 kDa or greater for complex polysaccharides, mucins, and other highly glycosylated proteins. Examples of biological products include, but are not limited to, vaccines, blood and blood components, allergens, somatic cells, gene therapies, tissues, and recombinant therapeutic proteins. In certain embodiments, biological products are drugs for treating diseases and / or medical conditions. Examples of biopharmaceuticals include, but are not limited to, natural or engineered antibodies or antigen-binding fragments thereof, and antibody-drug conjugates, which include antibodies or antigen-binding fragments thereof conjugated directly or indirectly (e.g., via a linker) to a drug of interest (such as a cytotoxic drug or toxin). In certain embodiments, biopharmaceuticals, such as gene therapy drugs, include vectors, such as adeno-associated virus (AAV), adenovirus, lentivirus, and retroviral vectors, and (e.g., loaded) nucleic acids, such as DNA or RNA. In certain embodiments, biological products are diagnostic agents for diagnosing diseases and / or medical conditions. For example, allergen patch testing utilizes biological products (e.g., biological products made from natural substances) known to cause contact dermatitis. Diagnostic biological products can also include medical imaging agents, such as proteins labeled with reagents (such as fluorescent markers, dyes, radionuclides, etc.) that provide a detectable signal to facilitate imaging.

[0311] In vitro: As used herein, the term "in vitro" refers to events that occur in an artificial environment, such as in a test tube or reaction vessel, in cell culture, etc., rather than within a multicellular organism.

[0312] In vivo: As used herein, the term "in vivo" refers to events that occur within a multicellular organism (such as humans and non-human animals). In the context of cell-based systems, the term can be used to refer to events that occur within living cells (as opposed to, for example, in vitro systems).

[0313] Peptide: As used herein, the term "peptide" refers to a polypeptide that is generally relatively short, for example having a length of less than about 100 amino acids, less than about 50 amino acids, less than about 40 amino acids, less than about 30 amino acids, less than about 25 amino acids, less than about 20 amino acids, less than about 15 amino acids, or less than 10 amino acids.

[0314] Polypeptide: As used herein, refers to a polymeric chain of amino acids. In some embodiments, the polypeptide has an amino acid sequence that occurs in nature. In some embodiments, the polypeptide has an amino acid sequence that does not occur in nature. In some embodiments, the polypeptide has an engineered amino acid sequence as it is designed and / or produced through the action of man. In some embodiments, the polypeptide can include natural amino acids, unnatural amino acids, or both, or be composed of natural amino acids, unnatural amino acids, or both. In some embodiments, the polypeptide can include only natural amino acids or be composed only of natural amino acids or include only unnatural amino acids or be composed only of unnatural amino acids. In some embodiments, the polypeptide can include D-amino acids, L-amino acids, or both. In some embodiments, the polypeptide can include only D-amino acids. In some embodiments, the polypeptide can include only L-amino acids. In some embodiments, the polypeptide can contain one or more side groups or other modifications, such as at the N-terminus of the polypeptide, at the C-terminus of the polypeptide, or any combination of modifications or attachments to one or more amino acid side chains. In some embodiments, such side groups or modifications can be selected from the group consisting of acetylation, amidation, lipidation, methylation, polyethylene glycolylation, etc., including combinations thereof. In some embodiments, the polypeptide can be cyclic and / or can include a cyclic moiety. In some embodiments, the polypeptide is not cyclic and / or does not include any cyclic moiety. In some embodiments, the polypeptide is linear. In some embodiments, the polypeptide can be or include a stapled polypeptide. In some embodiments, the term "polypeptide" can be appended to the name of a reference polypeptide, activity, or structure; in such cases, it is used herein to refer to polypeptides that share the relevant activity or structure and can thus be considered members of the same polypeptide class or family. For each such class, the present specification provides and / or those skilled in the art will know exemplary polypeptides within the class, the amino acid sequence and / or function of which is known; in some embodiments, such exemplary polypeptides are the reference polypeptides of the polypeptide class or family. In some embodiments, members of a polypeptide class or family show significant sequence homology or identity with the reference polypeptide of the class, share a common sequence motif (e.g., a characteristic sequence element) with the reference polypeptide of the class, and / or share a common activity (in some embodiments, at a comparable level or within a specified range) with the reference polypeptide of the class; in some embodiments, all polypeptides within the class).For example, in some embodiments, the member polypeptide exhibits an overall sequence homology or identity to a reference polypeptide of at least about 30-40%, and typically greater than about 50%, 60%, 70%, 80%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or higher, and / or comprises at least one region that exhibits very high sequence identity, typically greater than 90% or even 95%, 96%, 97%, 98% or 99% (e.g., a conserved region that can be or include a characteristic sequence element in some embodiments). Such conserved regions typically span at least 3-4 and typically up to 20 or more amino acids; in some embodiments, the conserved region spans at least a stretch of at least 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 or more contiguous amino acids. In some embodiments, the related polypeptide can comprise or consist of a fragment of the parent polypeptide. In some embodiments, a useful polypeptide can comprise or consist of multiple fragments, each fragment existing in the same parent polypeptide in a different spatial arrangement relative to each other as found in the polypeptide of interest (e.g., fragments that are directly linked in the parent can be spatially separated in the polypeptide of interest or vice versa, and / or the fragments can be present in a different order in the polypeptide of interest than in the parent), such that the polypeptide of interest is a derivative of its parent polypeptide.

[0315] Protein: As used herein, the term "protein" refers to a polypeptide (i.e., a string of at least two amino acids linked to each other by peptide bonds). A protein can contain moieties other than amino acids (e.g., can be a glycoprotein, proteoglycan, etc.) and / or can be otherwise processed or modified. One of ordinary skill in the art will appreciate that a "protein" can be a complete polypeptide chain produced by a cell (with or without a signal sequence), or can be a characteristic portion thereof. One of ordinary skill in the art will appreciate that a protein can sometimes contain more than one polypeptide chain, e.g., linked by one or more disulfide bonds or associated in some other manner. A polypeptide can contain L-amino acids, D-amino acids, or both, and can contain any of a variety of amino acid modifications or analogs known in the art. Useful modifications include, for example, terminal acetylation, amidation, methylation, etc. In some embodiments, a protein can include natural amino acids, unnatural amino acids, synthetic amino acids, and combinations thereof. The term "peptide" is generally used to refer to a polypeptide that is less than about 100 amino acids in length, less than about 50 amino acids in length, less than 20 amino acids in length, or less than 10 amino acids in length. In some embodiments, a protein is an antibody, an antibody fragment, a bioactive portion thereof, and / or a characteristic portion thereof.

[0316] Machine learning module, machine learning model: As used herein, the terms "machine learning module" and "machine learning model" are used interchangeably and refer to a computer-implemented process (e.g., a software function) that implements one or more specific machine learning algorithms, such as artificial neural networks (ANNs), convolutional neural networks (CNNs), random forests, decision trees, support vector machines, etc., in order to determine one or more output values for a given input. In some embodiments, for example, a dataset that is curated and / or manually annotated is used to train a machine learning module that implements machine learning techniques. Such training can be used to determine various parameters of the machine learning algorithm implemented by the machine learning module, such as the weights associated with the layers in a neural network. In some embodiments, once the machine learning module is trained, e.g., to perform a specific task such as determining the various metrics described herein, the values of the determined parameters are fixed and the (e.g., unchanging, static) machine learning module is used to process new data (e.g., different from the training data) and perform its trained task without further updating its parameters (e.g., the machine learning module does not receive feedback and / or updates). In some embodiments, the machine learning module can receive feedback, e.g., based on a user's review of the accuracy, and such feedback can be used as additional training data, e.g., to dynamically update the machine learning module. In some embodiments, the trained machine learning module is a classification algorithm with adjustable and / or fixed (e.g., locked) parameters, such as a random forest classifier. In some embodiments, two or more machine learning modules can be combined and implemented as a single module and / or a single software application. In some embodiments, two or more machine learning modules can also be implemented separately, e.g., as separate software applications. The machine learning module can be software and / or hardware. For example, the machine learning module can be fully implemented as software, or certain functions of the ANN module can be performed by dedicated hardware (e.g., by an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), etc.).

[0317] Mid-infrared, mid-IR, MIR: As used herein, the terms mid-infrared, mid-IR, and MIR are used interchangeably and refer to the portion of the electromagnetic spectrum that ranges from about 5000 cm -1 to about 500 cm -1 (corresponding to a wavelength range of about 2 μm to about 20 μm) and / or in certain embodiments from about 3,000 cm -1 to about 800 cm -1 (corresponding to a wavelength range of about 3 μm to about 12 μm).

[0318] Substantially: As used herein, the term "substantially" refers to a qualitative condition that exhibits all or nearly all of the range or degree of the characteristic or property of interest.

[0319] Detailed Description

[0320] In view of this, the systems, architectures, devices, methods, and processes of the present disclosure cover variations and adaptations developed using information from the embodiments described herein. As contemplated in this specification, adaptations and / or modifications of the systems, architectures, devices, methods, and processes described herein may be performed.

[0321] Throughout the specification, where an architecture, article, device, method, process, and system are described as having, including, or comprising particular components, or where a process and method are described as having, including, or comprising particular steps, in view of this, additionally, there are architectures, articles, devices, and systems of the present invention that consist essentially of or consist of the said components, and there are processes and methods according to the present invention that consist essentially of or consist of the said processing steps.

[0322] It should be understood that as long as the present invention remains operable, the order of steps or the order of performing certain actions is not important. In addition, two or more steps or actions may be performed simultaneously.

[0323] Any publication mentioned herein, such as in the background section, is not an admission that the publication is prior art with respect to any claim presented herein. The background section is presented for clarity purposes and is not meant to be a description of prior art with respect to any claim.

[0324] As noted, the literature is incorporated herein by reference. In the event of any discrepancy in the meaning of a particular term, the meaning provided in the above - defined section shall prevail.

[0325] Titles are provided for the convenience of the reader, and the presence and / or placement of a title are not intended to limit the scope of the subject matter described herein.

[0326] Methods and systems for monitoring and controlling the performance and operation of a production unit during various stages of a biomanufacturing process are described herein. In certain embodiments, the bioproduction monitoring and / or control techniques described herein utilize a mid - infrared (MIR) analyzer to measure an infrared (IR) absorption signal from a liquid sample and generate IR absorption data, such as an IR spectrum, substantially in real - time. The liquid sample measured in this manner can be used as an input and / or output of one or more production units used in biologic manufacturing.

[0327] As described in further detail herein, mid-infrared spectroscopy provides a powerful non-destructive and label-free analytical technique that facilitates the rapid and accurate identification and characterization of liquid samples, including biological materials. In certain embodiments, the bioproduction monitoring and control techniques described herein use IR spectroscopic data to determine sample quality metrics that, in particular, provide a measure of sample characteristics, such as the content of one or more desired target molecules, the presence of impurities, molecular structure information, and the like. One or more sample quality attributes (e.g., critical quality attributes) can be monitored in real time and used individually and / or in combination with each other and / or in combination with data from other sensors to control and / or improve the operation of one or more production units, thereby facilitating compliance with regulations and / or meeting the demanding quality tolerances that may be required to produce increasingly effective and / or safe products, expanding production capacity, increasing efficiency, and the like.

[0328] A. Mid-infrared (MIR) spectroscopy

[0329] Go to Figure 1 A, mid-infrared spectroscopy can be used to obtain detailed information about the vibrational transitions of biomolecules such as carbohydrates, lipids, nucleic acids, proteins, etc. As Figure 1B shown, mid-infrared spectroscopy measures the IR absorption signal generated by the vibrational modes of molecules. When a molecule is irradiated with mid-infrared light (including a range of MIR frequencies / wavelengths), it can absorb the light to varying degrees at various frequencies / wavelengths. Molecules have a characteristic set of vibrational modes that depend, in particular, on their molecular structure and local environment. Thus, as Figure 1 shown in A, various molecules such as lipids, proteins, nucleic acids, carbohydrates, etc. absorb light within one or more characteristic bands. Absorption at these characteristic bands can be observed, for example, as a series of peaks in an IR absorption spectrum. As described in further detail herein, the characteristics of these peaks, such as their amplitude, linewidth, center frequency, area, etc., can be used to determine metrics for measuring sample characteristics, such as total protein content, molecular identity, and / or heterogeneity, etc.

[0330] A.i Mid-infrared spectral bands

[0331] In certain embodiments, one or more mid-infrared spectral bands are associated with nucleic acid molecules. For example, as Figure 2 shown, one or more peaks in the mid-infrared spectrum can be associated with the asymmetric PO4 stretching band and / or the symmetric PO4 stretching band, which for the Figure 2 sample shown are present at 1220 - 1250 cm -1 and 1075 - 1100 cm -1nearby and are thus used to detect the presence and / or characterize nucleic acids (such as DNA and / or RNA) within a sample. As described in further detail herein, IR absorption bands associated with the asymmetric PO4 stretching vibration mode can occur in the range of approximately 1150 cm -1 to approximately 1250 cm -1 (e.g., in the range of approximately 1175 cm -1 to approximately 1250 cm -1 ; e.g., in the range of approximately 1200 cm -1 to approximately 1250 cm -1 ; e.g., in the range of approximately 1210 cm -1 to approximately 1230 cm -1 ). IR absorption associated with the symmetric PO4 stretching vibration mode can occur in the range of approximately 1000 cm -1 to approximately 1100 cm -1 (e.g., in the range of approximately 1050 cm -1 to approximately 1100 cm -1 ; e.g., in the range of approximately 1075 cm -1 to approximately 1100 cm -1 ; e.g., in the range of approximately 1075 cm -1 to approximately 1085 cm -1 ). As Figure 2 shown, in certain embodiments, the characteristics (such as center frequency, shape, individual and / or relative amplitude) of the peaks associated with these two (symmetric and asymmetric) PO4 stretching bands can be used to characterize the type and / or specific conformation of nucleic acid molecules. For example, Figure 2 shows the variation of these spectral peaks for several nucleic acid molecules (including single-stranded and double-stranded DNA and RNA).

[0332] In certain embodiments, one or more mid-infrared spectral bands can be associated with proteins and used to evaluate protein properties. For example, as Figure 1 shown in A, the amide I and amide II bands can be associated with spectral peaks in the range of approximately 1500 to approximately 1700 cm -1 and can be used to detect and / or characterize the protein content and / or structure within a sample. For example, without wishing to be bound by any particular theory, it is believed that the amide-I and amide-II bands correspond to vibration modes associated with the atoms of the protein peptide backbone. Thus, in certain embodiments, the presence and intensity of these (amide-I and amide-II) bands can be used to determine the presence and / or content of protein in a sample. In certain embodiments, additionally or alternatively, other bands associated with protein (e.g., backbone) vibrations, such as the amide-III band in the range of approximately 1250 cm -1 to approximately 1350 cm -1 , can be used to determine the presence and / or content of protein in a sample.

[0333] Additionally or alternatively, in some embodiments, amide-I band measurements can be used to characterize the secondary structure of one or more proteins in a sample and / or changes therein. For example, without wishing to be bound by any particular theory, it is believed that the amide I band originates from molecular vibrations (C=O stretching vibrations) in the protein peptide backbone and is sensitive to the protein secondary structure (e.g., conformation). For example, as Figure 3 shown, protein secondary structure motifs such as random coils, α-helices, and β-sheets each produce characteristic amide-I absorption peaks with characteristic central frequencies, line widths, and splittings.

[0334] Regarding protein measurements, in some embodiments, mid-infrared spectroscopy techniques offer several advantages over ultraviolet (UV) absorption measurements. Turning to Figure 4 and 5 , where UV absorption measures absorption at or near 280 nm, which originates from three specific amino acids - tyrosine (Tyr), tryptophan (Trp), and phenylalanine (Phe). Thus, not all amino acids of a protein produce a measurable UV absorption signal. In contrast, all amino acids (linking their side chains and / or peptide bonds) can contribute to detectable absorption in the IR, which can be measured by mid-infrared spectroscopy. For example, the amide-I, amide-II, and amide-III absorption features are associated with the vibrational modes of the protein backbone atoms, and thus, all proteins produce said absorption features and are roughly proportional to the amino acid count (intensity, i.e., absorption level). Without wishing to be bound by any particular theory, it is thought that the amide-I band is (primarily) produced by C=O bond vibrations, and it is thought that amide-II and III are produced by NH and CN bond vibrations (out-of-phase and in-phase combinations, respectively). Thus, mid-infrared-based measurements are not limited to specific molecular weights and / or protein types. Additionally, UV absorption has poor linearity and cannot be used to evaluate (e.g., deconvolute) the heterogeneity of protein mixtures. As described and demonstrated herein, mid-infrared absorption spectroscopy can be used in particular to measure total protein concentration and, additionally or alternatively, to quantify the heterogeneity of protein mixtures, thereby providing functionality and insights into protein samples not achievable with UV absorption measurements.

[0335] A.ii Mid-Infrared Analyzer

[0336] In certain embodiments, the production monitoring and / or control techniques described herein utilize one or more MIR analyzers to measure infrared (IR) absorbance signals from a sample mixture and generate IR absorbance data, which can be analyzed, for example, to identify and / or characterize spectral peaks associated with characteristic molecular absorption bands and / or their peak characteristics. A mid-infrared analyzer can be or include a system that includes one or more components such as a mid-infrared light source, a detector, and associated optics, such as (but not limited to) sampling optics, to direct light to and / or from a sample in a specific manner to interrogate it through a specific sampling geometry.

[0337] MIR light sources

[0338] In certain embodiments, the MIR analyzer includes one or more MIR light sources operable to emit MIR light. For example, in certain embodiments, the MIR light source is or includes a heat source that emits light including a wide range of wavelengths, the spectral profile of which generally corresponds to a blackbody spectrum at a specific temperature, the specific temperature being selected, for example, to place most of its emission power within the MIR.

[0339] In certain embodiments, one or more light sources of the MIR analyzer include one or more lasers that emit light within the MIR at substantially a single frequency (e.g., within a narrow band near a center wavelength). The MIR laser can be a tunable laser such that its emission frequency can be tuned within a specific spectral range. Examples of tunable MIR lasers include, but are not limited to, quantum cascade lasers (QCLs). Other laser-based sources can include, but are not limited to, interband cascade lasers (ICLs), sources based on difference frequency generation, frequency combs (e.g., dual-comb sources), etc. As described in further detail herein, sources based on tunable QCLs provide various advantageous features such as high spectral brightness and a flexible tuning range within key spectral windows in the mid-infrared.

[0340] Spectral measurement

[0341] In certain embodiments, the MIR analyzer can measure IR absorption signals at multiple wavelengths and / or times to generate IR spectral data that provides a measure of detection signals and / or absorbance at multiple wavelengths within a specific spectral range. Fig. 6A -F shows and compares two methods for performing IR spectral measurements. One method is an interferometric technique called Fourier transform infrared (FT-IR) spectroscopy. As Fig. 6A shown, FT-IR spectroscopy uses a beam-splitting interferometer with a moving mirror to record an interference pattern that varies with the time delay between two light beams. The interference pattern obtained in this way can then be Fourier-transformed to obtain an IR spectrum. As Figure 6C and 6E As shown, FT-IR spectroscopy irradiates a sample at multiple frequencies simultaneously and thus is typically used with a broadband incoherent source (such as the heat source described herein).

[0342] In some embodiments, the MIR analyzer uses a spectral scanning technique to record the IR spectrum. As Figure 6B shown, a tunable IR laser (such as a tunable QCL) can be used to implement the spectral scanning method. Turning to Fig.6D and 6F , the tunable laser emits light within a narrow band at substantially a single specific emission frequency / wavelength. The emission frequency of the tunable laser can be scanned to irradiate the sample at multiple wavelengths one at a time within the tuning range of the tunable laser. Then the signal can be detected at each of the multiple scanned wavelengths to build the IR spectrum, one wavelength at a time.

[0343] Some Advantages of QCL-Based Mid-IR Spectroscopy

[0344] The MIR laser source can offer advantages over the heat source. In addition to this, as Figure 7 shown, a laser source such as a QCL emits a strong MIR beam, the spectral brightness of which (in units of W×sr -1 ×m -2 ×μm -1 , e.g., watts per square meter per steradian per unit wavelength) is several orders of magnitude greater (about 10 4 -10 6 times) than that of the heat source. Figure 7 Three QCLs are shown, each having a tuning range of about 2 - 3 μm.

[0345] Turning to Figure 8 , in some embodiments, the high spectral brightness provided by a laser source such as a QCL eliminates several significant drawbacks that have historically limited the application of conventional IR spectroscopic instruments (such as FTIR), which rely on heat sources, for measuring biological samples and processes, particularly in aqueous environments.

[0346] First, as Figure 8 shown, liquid water has two strong and broad absorption bands in the MIR, one of which (at about 1638 cm -1The H-O-H bending mode centered around it overlaps substantially with the amide-I and amide-II bands used to measure protein content and characterize the structure. Therefore, IR measurements in water are limited to flow cells with extremely short path lengths, which are not compatible with on-line monitoring of biological process workflows. Secondly, due to their low brightness, heat sources are typically used with cryogenically cooled (e.g., with liquid nitrogen) detectors (e.g., MCT detectors) in order to obtain sufficient sensitivity. However, such cryogenically cooled detectors are cumbersome to operate and have poor linearity, long-term stability, and reproducibility. Thirdly, low spectral brightness heat sources require long data acquisition times in order to achieve sufficient sensitivity. These (long acquisition times) are incompatible with PAT requirements.

[0347] The high spectral brightness provided by MIR laser sources (such as QCLs) overcomes these historical limitations of IR spectroscopy, thus allowing much longer path length measurements in water, the use of electronically cooled detectors, and fast acquisition times.

[0348] Mid-infrared detector

[0349] In certain embodiments, the MIR analyzer includes one or more detectors. A variety of detectors operable to detect light in the MIR range can be used to detect MIR light, for example as part of an MIR analyzer. The detector can be a single-element detector or a multi-element detector, such as a linear array or a focal plane array (FPA). In certain embodiments, the MIR detector is cryogenically cooled, for example by liquid nitrogen. In certain embodiments, the MIR detector is thermoelectrically cooled or uncooled. In certain embodiments, the MIR detector is a quantum detector, such as a mercury cadmium telluride (MCT) detector. In certain embodiments, the MIR detector is a thermal detector, such as a deuterated L-alanine doped triglycine sulfate (DLaTGS) or deuterated triglycine sulfate (DTGS) detector. In certain embodiments, the MIR detector is a bolometer or a microbolometer.

[0350] In certain embodiments, the MIR analyzer includes a single detector. In certain embodiments, the MIR analyzer includes two or more detectors. For example, the MIR analyzer (such as the MIR analyzer shown in Figure 6B may include a sample detector that detects the signal of light that has passed through the sample, and a reference detector that detects the signal of a portion of the illumination beam that is split (e.g., by a beam splitter) before the sample. Among them, the sample detector and the reference detector can be used in this way to compensate for laser power fluctuations.

[0351] Sampling optics and geometry

[0352] In some embodiments, the MIR analyzer includes sampling optics for directing an MIR beam from an MIR source onto and / or into a specific region of a sample and then to a detector for detecting a signal from the sample region. The sampling optics can be used to interrogate the sample in a specific manner, such as, but not limited to, by providing a transparent window through which the beam can pass, by directing light onto / into a region of the sample at a specific angle, such as through reflective elements (e.g., mirrors and high refractive index materials), and by focusing the beam using lenses or curved (e.g., parabolic) reflectors.

[0353] For example, the sampling optics can be used to interrogate the sample using a specific type of sampling geometry, such as transmission or attenuated total internal reflection (ATR) geometry. In some embodiments, the transmission geometry uses the sampling optics to direct an infrared beam along a substantially straight path through the sample and then to a detector after passing through the sample. In this way, the transmission geometry measures the absorption of IR light generated by its propagation through the sample.

[0354] In some embodiments, the ATR geometry uses sampling optics that include a high refractive index material (such as an ATR crystal) whose surface is in contact with the sample. The sampling optics direct the infrared beam into the high refractive index material such that it impinges on the surface in contact with the sample at an angle greater than the angle required for total internal reflection (TIR angle). Thus, the light is reflected by the high refractive index material - sample interface and returns to the detector, probing the sample with an evanescent wave rather than propagating through it. The penetration depth of the evanescent wave and thus the effective path length within the sample can be controlled by varying the angle of incidence at the interface between the high refractive index material and the sample and / or by selecting a specific material as the high refractive index material. In some embodiments, the ATR geometry uses an ATR crystal that has a shape including one or more angled surfaces through which light can enter and / or exit the ATR crystal and a flat surface in contact with the sample. Examples of high refractive index materials that can be used for the ATR crystal include, but are not limited to, diamond, germanium (Ge), silicon (Si), thallium halides (e.g., thallium bromide, also known as KRS-5), and zinc selenide (ZnSe). In some embodiments, the ATR crystal is a multi-reflection ATR crystal that is shaped to cause the light to reflect multiple times within the crystal in order to increase the effective path length for probing the sample. In some embodiments, the ATR crystal is located at the end of an optical fiber probe. In some embodiments, the optical fiber probe itself can be used as the high refractive index material to achieve the ATR sampling geometry. Examples of high refractive index materials for IR optical fiber probes include, but are not limited to, chalcogenides, silver halides, fluoride glasses (such as ZBLAN), etc.

[0355] Flow Cell

[0356] In some embodiments, the MIR analyzer includes a flow cell. According to various embodiments, any flow cell suitable for the application can be used. Non-limiting examples of MIR analyzers with flow cells are described in detail in, for example, U.S. Patent No. 10,753,856, issued August 25, 2020, U.S. Patent Publication No. 2021 / 0405001A1, issued December 30, 2021, and U.S. Patent No. 11,119,079, issued September 14, 2021, the respective contents of which are incorporated herein by reference in their entirety.

[0357] Exemplary QCL-based MIR spectrometer

[0358] Various MIR analyzers based on QCL sources are described in detail in, for example, U.S. Patent No. 10,753,856, issued August 25, 2020, U.S. Patent Publication No. 2021 / 0405001A1, issued December 30, 2021, and U.S. Patent No. 11,119,079, issued September 14, 2021, the respective contents of which are incorporated herein by reference in their entirety.

[0359] Fig. 9A An exemplary QCL-based MIR analyzer for absorption measurements in solution is shown. The QCL-based MIR analyzer includes a tunable QCL laser source. As shown in inset 904, the QCL source is a scanning source that repetitively scans its emission wavelength over a specific tuning range, completing a full spectral scan (i.e., across the entire tuning range) approximately once per second. Turning to Fig. 9B , different QCL sources can have different tuning windows and can thus be used to probe different parts of the MIR spectral range.

[0360] For example, as Fig. 9B shown, a commercial QCL-based IR spectrometer (Culpeo-LA-P from Daylight Solutions) has a spectral window ranging from approximately 1725 cm -1 to approximately 1375 cm -1 , which can be used, for example, to measure amide bands related to proteins and can thus be used for protein detection and characterization. Another QCL (e.g., Daylight Culpeo-LA-S) can have a different spectral scan window, such as approximately 1375 cm -1 to approximately 1025 cm -1 or approximately 1225 cm -1 to approximately 1000 cm -1 , which contains bands related to sugars, polysaccharides, and nucleic acids. In some embodiments, multiple QCL sources can be used to cover the desired spectral range. For example, two QCL sources can be combined to cover approximately 1725 cm-1 to about 1025 cm -1 In certain embodiments, two or more QCL sources can be included in a single MIR analyzer such that they share at least a portion of the sampling optics and / or detector. In certain embodiments, two separate MIR analyzers can be used to provide the desired spectral coverage, each MIR analyzer having a specific QCL with a specific tuning window.

[0361] Commercial implementations of exemplary QCL-based IR spectrometers provide various performance characteristics that are advantageous for mid-infrared spectroscopy-based measurements of bioproduction processes. For example, a QCL-based IR spectrometer can allow quantitative measurements to be made substantially in real time, e.g., at a rate of about 1 Hz, over a wide dynamic range (e.g., from below 0.1 to above 300 mg / mL; e.g., from about 0.001 to above 300 g / L). In certain embodiments, a QCL-based IR spectrometer is compatible with flow rates up to about 10 L / min and / or can detect sample volumes as small as picoliters. In certain embodiments, a QCL-based IR spectrometer is a modular instrument and / or is suitable for in-line and / or at-line measurements.

[0362] Table 1 below shows the advantages of a QCL-IR spectrometer system compared to other techniques for measuring bioproduction processes:

[0363] Table 1: Certain Advantages and Challenges of Process Monitoring Techniques

[0364]

[0365] A.iii Absorbance Data and Preprocessing

[0366] In certain embodiments, the IR absorption signals measured from a sample are mathematically combined and / or preprocessed to produce an IR absorbance spectrum indicative of the sample absorbance. For example, Fig. 10A -H shows the various steps and mathematical processing for obtaining the absorbance spectrum of an analyte present in a mobile phase. Wherein, as Fig. 10A -H shows, one or more reference spectra of the mobile phase indicative of the absence of the analyte can be measured and divided by / subtracted from to produce a spectrum of the analyte present in the sample including the mobile phase.

[0367] B. Real-time bioproduction monitoring and process control

[0368] In certain embodiments, the bioproduction monitoring and / or control techniques described herein utilize analytical techniques, particularly mid-infrared spectroscopy as described herein, for feedstock analysis, process monitoring and control, and end-product analysis. In certain embodiments, the techniques described herein are implemented as part of a Process Analytical Technology (PAT) framework, such as utilizing mid-infrared spectroscopy as an integrated component within a bioprocess workflow to, inter alia, identify sources of variability, monitor and facilitate the management of these sources of variability (e.g., via system control and feedback), and ensure that product quality attributes (e.g., critical quality attributes) can be accurately and reliably predicted over the design space established for the materials used, process parameters, manufacturing, environment, and other conditions. The techniques described herein can, for example, be used as a process fingerprinting tool in bioprocess unit operations. In certain embodiments, the techniques described herein can additionally or alternatively be used, for example, in biopharmaceutical forensic laboratories to detect counterfeit drugs and biosimilars.

[0369] In certain embodiments, one or more mid-infrared analyzers can be used as in-line sensors embedded within the process stream to monitor sample quality attributes substantially in real time. Mid-infrared-based monitoring can be used alone and / or in combination with other measurement modalities. IR spectral data can be processed by a variety of methods, including methods for determining specific metrics associated with certain absorption bands, as well as advanced methods in machine learning techniques. Accordingly, sample quality metrics can be monitored and / or used to control the bioproduction process. In certain embodiments, automated and / or semi-automated decision support and / or control systems (including, for example, artificial intelligence (AI)-based systems) can be used for real-time and / or refined process control.

[0370] B.i. Mid-infrared-based Monitoring of Sample Quality Metrics

[0371] Sample quality metrics

[0372] In some embodiments, IR absorption data obtained from a liquid sample can be used to determine one or more sample quality metrics that characterize one or more target analytes within the sample. In some embodiments, a sample quality metric is or includes a value and / or a set of values that characterize one or more properties (e.g., physical, chemical, biological, or microbiological properties) of one or more target analytes in the sample. In some embodiments, the sample quality metric characterizes the properties of one or more bioanalytes such as proteins, viruses and / or virus-like particles, nucleic acids, etc. In some embodiments, the target analyte is a desired class of biologic to be purified and retained, such as a specific protein, viral vector, nucleic acid, or form thereof. In some embodiments, the target analyte is an undesired impurity, such as a portion of a mixture to be removed. A sample quality metric can include, but is not limited to, specific attributes (referred to as “critical quality attributes (CQAs)”) that should be within appropriate limits, ranges, or distributions to ensure a desired product quality.

[0373] A sample quality metric can be a value, such as a numerical value, that provides a measure of the amount of a specific target analyte (such as a specific molecular species or form thereof) in the sample. For example, the numerical value can be a direct measurement of a physical amount, such as total mass, number, concentration, etc., or can be a value that is proportional to, indicative of a relative change in, or related to the physical amount of a specific target analyte. In some embodiments, a sample quality metric can be a measure that indicates a specific species or state based on whether its value lies within one or more specific (e.g., pre-specified) ranges and / or is above or below one or more thresholds.

[0374] The amount of a specific analyte and / or its subspecies. In some embodiments, a sample quality metric is or includes a measure of the amount of a specific target analyte, such as the concentration, total mass, etc. of one or more target molecules. A sample quality metric can include, for example, real-time measurements of the total protein concentration (e.g., titer), total protein mass, etc. in the sample. Additionally or alternatively, a sample quality metric can include, for example, real-time measurements of the total nucleic acid concentration (e.g., titer), total nucleic acid mass, etc. in the sample. In some embodiments, a sample quality metric includes (e.g., real-time) content measurements of a target analyte, such as the total concentration, mass, etc. of one or more of the following: lipids, polysaccharides, etc. In some embodiments, a sample quality metric includes (e.g., real-time) content measurements of the assembly of multiple molecules, such as the total concentration (e.g., titer), total mass, number (e.g., discrete number) of viruses and / or virus-like particles (such as viral vector assembly, including but not limited to adenovirus, adeno-associated virus (AAV), retrovirus (e.g., lentivirus), plant-based virus (e.g., tobacco mosaic virus), etc.).

[0375] In certain embodiments, a sample quality metric can be or include a measure of the absolute and / or relative amount of a particular type or form of molecule (e.g., a biomolecule such as a protein, nucleic acid, lipid, polysaccharide, etc.). For example, in certain embodiments, one or more sample quality metrics can be or include a measure of molecular conformation and / or heterogeneity, such as protein secondary structure, aggregation, the identity and relative concentration of various protein species, conjugation (e.g., glycosylation), antibody-drug conjugate ratio, etc.

[0376] For example, in certain embodiments, a sample quality metric can be a measure of the absolute or relative amount of a particular protein secondary structure motif. For example, as described in further detail herein, IR spectroscopy can be used to identify the content of protein secondary structure motifs such as α-helices, β-sheets, β-turns, and disordered secondary structures. Thus, in certain embodiments, a sample quality metric can be a measure, such as a numerical value proportional to or related to the content of a particular secondary structure motif. In certain embodiments, a sample quality metric can be a measure of relative amount, such as a measure of the relative amount between two secondary structure motifs.

[0377] For example, in certain embodiments, a sample quality metric can be or include a measure of the absolute and / or relative amount of a particular form of protein resulting from one or more post-translational modifications, such as covalent addition of a functional group or protein, proteolytic cleavage of a regulatory subunit, or degradation of the entire protein, e.g., phosphorylation, glycosylation, ubiquitination, nitrosylation, lipidation, and proteolysis, etc.

[0378] In certain embodiments, a sample quality metric can be or include a measure of the absolute and / or relative amount of a particular nucleic acid conformation, such as the concentration (e.g., titer), total mass, etc. of single-stranded (ss) DNA, double-stranded (ds) DNA, B DNA, A DNA, Z DNA, triple-helix DNA, etc., and / or the relative fraction of any of the foregoing, e.g., with respect to (e.g., relative to) the total DNA, nucleic acid, etc. content. In certain embodiments, a sample quality metric can be or include a measure of the absolute and / or relative amount of one or more particular types of nucleic acid bases (e.g., guanine (G), cytosine (C), thymine (T), adenine (A), uracil (U), etc.) and / or combinations thereof. For example, in certain embodiments, a sample quality metric can be or include a GC content metric that provides a measure of the total and / or relative amount of GC bases within a sample. For example, a GC content metric can be or include a measure of the total GC content in a sample, such as the concentration, total mass, etc. of GC. In certain embodiments, a GC content metric can be or include a measure of relative GC content, e.g., scaled relative to the total nucleic acid content, or e.g., incorporating a known, expected, or assumed value, such as strand length, to provide an average measure of the GC content per nucleic acid molecule / strand (e.g., the average percentage of GC bases per nucleic acid molecule in a sample).

[0379] In some embodiments, the sample quality metric can be or include a value indicative of the degree of aggregation or its absence and / or presence in the sample. For example, the sample quality metric can be a value (e.g., a numerical value) that provides a measure of the content of a particular type of aggregate (such as monomers, dimers, trimers, multimers, etc.) in the sample. For example, a sample quality metric can be a value that provides a measure of the content of a particular type of protein aggregate within the sample, e.g., proportional and / or related (e.g., increasing or decreasing) to the content of a particular type of protein aggregate within the sample. For example, one sample quality metric can measure the monomer content within the sample, and another sample quality metric can measure the dimer and / or higher molecular weight species (e.g., dimers, trimers, etc.) content within the sample. In some embodiments, the sample quality metric can indicate a particular aggregate species or state based on a comparison of its value to one or more ranges and / or thresholds. For example, in some embodiments, as described in further detail herein, the sample quality metric can indicate a monomeric protein species when its value falls within a particular range, and the presence of aggregation (e.g., dimers and / or multimers) when its value moves outside of the particular range. The particular range can be a pre-specified numerical range, can be calibrated for a particular sample or protein species, or can be determined in real time, e.g., based on measurements during a particular sample processing run, such as chromatographic elution.

[0380] Identification of an analyte. In some embodiments, the sample quality metric can be or include a value that identifies the presence or absence of a particular target analyte (e.g., or a sufficient amount thereof) within the sample. For example, in some embodiments, the sample quality metric can be or include a Boolean value that has two states (e.g., 1 or 0, true or false, etc.) indicating the presence or absence of a particular target analyte within the sample, such as a desired protein or protein species and / or an undesired impurity. In some embodiments, the sample quality metric is or includes a value that identifies one or more particular analytes within the sample. For example, the sample quality metric can be or include a value or set of values that encode the identity of one or more components within the sample, such as an alphanumeric string, a set of strings, and / or alphanumeric characters, numbers, or Boolean arrays, etc.

[0381] In some embodiments, the sample quality metric can relate to gene therapy products. For example, in some embodiments, the sample quality metric can characterize the content of viral particles, nucleic acid content, and / or mixtures or assemblies thereof. In some embodiments, the sample quality metric can be or include a measured value of the virus and / or capsid titer as described herein. In some embodiments, the sample quality metric can be or include a measure of the capsid content, such as the total content of empty capsids, the total content of full capsids, or a relative measure, such as the fraction, percentage, etc. of empty to full capsids.

[0382] Certain mid-infrared spectral bands for determining sample quality metrics. In certain embodiments, mid-infrared absorbance data (such as mid-infrared spectral data) is used to determine sample quality metrics. In certain embodiments, mid-infrared spectral data including an amide I region and / or an amide II region is used to determine sample quality metrics, where the amide I region ranges from about 1600 cm -1 to about 1700 cm -1 or about 1800 cm -1 (e.g., ranging from about 1600 cm -1 to about 1725 cm -1 ; e.g., ranging from about 1625 cm -1 to about 1725 cm -1 ; e.g., ranging from about 1630 cm -1 to about 1650 cm -1 ), and the amide II region ranges from about 1500 to about 1600 cm -1 (e.g., ranging from about 1500 cm -1 to about 1575 cm -1 ; e.g., ranging from about 1500 cm -1 to about 1550 cm -1 ; e.g., ranging from about 1540 cm -1 to about 1560 cm -1 ). In certain embodiments, mid-infrared spectral data including one or both of the amide I and amide II spectral regions can be used to determine one or more sample quality metrics indicative of and / or characterizing one or more protein species within a sample. In certain embodiments, mid-infrared spectral data having a spectral range from about 1000 cm -1 to about 1350 cm -1 is used to determine sample quality metrics, which can, for example, be used to determine one or more sample quality metrics indicative of and / or characterizing viruses and / or nucleic acid species (e.g., DNA, mRNA, etc.) within a sample. In certain embodiments, mid-infrared spectral data having a spectral range from about 1000 cm -1 to about 1700 cm -1 (e.g., up to about 1800 cm -1 ) is used to determine sample quality metrics. For example, in certain embodiments, a system capable of measuring from about 1000 cm -1 to up to 1800 cm -1 can be used to monitor a combined or multi-component process stream containing virus and / or protein (e.g., monoclonal antibody) species.

[0383] In certain embodiments, multiple spectral bands are used to determine a sample quality metric, e.g., to quantify protein and nucleic acid content within a sample. In certain embodiments, measurements of protein and nucleic acid content within a sample can be combined to additionally or alternatively determine a sample quality metric that measures viral capsid content and / or full / empty capsid content and / or relative fractions.

[0384] For example, Fig.11 An illustrative schematic diagram of an IR absorption spectrum of a sample comprising ssDNA and protein is shown, including a composite spectrum 1102 corresponding to the raw spectrum obtained from a sample comprising both ssDNA and protein, and spectra corresponding to the individual protein 1104 and ssDNA 1106 components. As described herein, protein content can be measured using the amide I 1112 and / or amide II 4014 spectral bands. Additionally or alternatively, in certain embodiments, protein content can be measured using IR absorption data within the amide-III spectral region 1116, which ranges from about 1250 cm -1 to about 1350 cm -1 (e.g., ranging from about 1250 cm -1 to about 1325 cm -1 ; e.g., ranging from about 1275 cm -1 to about 1325 cm -1 ; e.g., ranging from about 1280 cm -1 to about 1300 cm -1 ).

[0385] Nucleic acid content can be quantified using IR absorption data within a spectral range associated with asymmetric and / or symmetric phosphate stretching (PO4) vibrations. For example, in certain embodiments, a sample quality metric that measures nucleic acid content (e.g., a nucleic acid content metric) can be determined using IR absorption data from the asymmetric-PO4 spectral band 1118 (e.g., ranging from about 1150 cm -1 to about 1250 cm -1 ; e.g., ranging from about 1175 cm -1 to about 1250 cm -1 ; e.g., ranging from about 1200 cm -1 to about 1250 cm -1 ; e.g., ranging from about 1210 cm -1 to about 1230 cm -1 )(shown in Fig.11 as peaking at about 1220 cm -1 ). In certain embodiments, a sample quality metric that measures nucleic acid content (e.g., a nucleic acid content metric) can be determined using IR absorption data from the symmetric-PO4 spectral band 1120 (e.g., ranging from about 1000 cm-1 to about 1100 cm -1 ; for example, the range is from about 1050 cm -1 to about 1100 cm -1 ; for example, the range is from about 1075 cm -1 to about 1100 cm -1 ; for example, the range is from about 1075 cm -1 to about 1085 cm -1 ) to determine (shown in Fig.11 to peak at about 1080 cm -1 ).

[0386] In certain embodiments, specific spectral bands can be used / selected to allow independent quantification of the protein and nucleic acid content in a complex sample containing both proteins and nucleic acids. For example, as Fig.11 shown, the absorption within the amide-I spectral region 1112 of the spectrum taken from mixture 1102 can be contributed by both the protein 1104 and nucleic acid 1106 components, while the amide-II 4014 and amide-III 1116 absorptions are mainly due to Fig.11 shown in the protein 1104, and the ssDNA spectrum 4006 is relatively flat / minimal in these two (amide-II and amide-III) regions. Thus, in certain embodiments, the amide-I and / or amide-III bands can be used to quantify the protein content in samples where nucleic acids are present or may be present. The antisymmetric and symmetric PO4 regions 1118 and 1120 (where absorption is mainly generated by the nucleic acid content) can be used to quantify the nucleic acid content, for example, independent of changes in protein content.

[0387] Peak measurement

[0388] In certain embodiments, the values of one or more sample quality metrics described herein can be determined and monitored by analyzing one or more absorption bands within mid-infrared spectral data. In particular, in certain embodiments, the values of one or more peak metrics are determined for each of one or more specific absorption bands. Peak metrics are intended to quantify the characteristics of absorption bands, such as frequency position (e.g., center frequency), linewidth, intensity of one or more absorption bands, and can be calculated by various methods.

[0389] Frequency position metric. In certain embodiments, the peak metric is or includes a measure of the frequency position of a specific absorption band. For example, in certain embodiments, the frequency position (e.g., center frequency) of a specific absorption band can be or include the peak frequency (ν Peak) which can be determined by identifying the frequency at which the amplitude of a particular absorption band peaks (e.g., reaches a maximum), by fitting a predefined function (such as a Gaussian or Lorentzian function) and obtaining the center frequency from the fitted function, or by other methods directly from the absorption spectrum. In some embodiments, the frequency position (e.g., center frequency) of a particular absorption band can be or include the centroid frequency (ν COM )。In some embodiments, ν of a particular absorption band can be calculated from an absorption spectrum (e.g., a mid-infrared absorption spectrum) A(ν) according to Equation 1 below COM 。

[0390] (Equation 1)

[0391] where ν min and ν max are the lower (minimum frequency) and upper (maximum frequency) limits of the particular absorption band.

[0392] Line width metric. In some embodiments, one or more peak metrics can be or include a line width measurement, such as full width at half maximum (FWHM), which is calculated, for example, by various techniques, such as descending from the peak along one or both sides, or by fitting a function (e.g., Gaussian, Lorentzian, etc.).

[0393] Peak intensity metric. In some embodiments, the peak metric can be or include a measure of the intensity of one or more particular absorption bands, such as peak amplitude or area under the curve (AUC). In some embodiments, the peak amplitude of a particular band is calculated from the absorbance spectrum A(ν), which is the value at a particular (e.g., nominal) center frequency ν band associated with the particular band, i.e., A band = A(ν band ). In some embodiments, the peak amplitude of a particular band is calculated from the absorbance spectrum A(ν), which is the value at the peak frequency ν Peak i.e., A band = peak(Band) = A(ν Peak ). In some embodiments, a measure of the intensity of one or more particular absorption bands is determined by calculating the AUC. For a particular absorption band, the AUC can be calculated by integrating the absorbance spectrum A(ν) over the boundaries of the particular absorption band. For example, for a particular absorption band ranging from ν min to ν max , the AUC can be calculated according to Equation (2) below:

[0394] (Equation 2)

[0395] In some embodiments, the AUC of a single spectral band is calculated. In some embodiments, the AUC of two or more spectral bands is calculated, such as two or more adjacent bands or a region including multiple spectral bands that are considered to be associated with a particular target analyte of interest. In such cases, the AUC of a set of two or more spectral bands or a particular overall region can be calculated according to equation (2) above, where ν min and ν max specify the boundaries of two or more spectral bands or a particular overall region.

[0396] Combined peak metrics. In some embodiments, the sample quality metric can be a particular peak metric or calculated from a particular peak metric (e.g., as a function of a particular peak metric). For example, in some embodiments, the sample quality metric can be a particular peak metric, such as a measure of the spectral band (frequency) position. In some embodiments, the sample quality metric can be determined from a combination (e.g., calculated as a function thereof) of two or more (e.g., two; e.g., three or more; e.g., multiple) peak metrics, which can be from the same and / or different absorption bands. For example, in some embodiments, the sample quality metric can be determined as the sum, difference, ratio, product, or other function of two or more peak metrics. In some embodiments, the sample quality metric can be calculated as a function of two or more peak metrics and other values or constants (such as scaling factors, normalization constants, etc.), which can be known a priori and / or assumed and / or determined, for example, by molecular structure or other known properties and / or from other (e.g., orthogonal) measurement methods, such as other spectroscopic techniques, previous measurements, etc., as described herein. For example, in some embodiments, the sample quality metric can be determined as a linear function or combination of one or more peak metrics, for example having the form shown in equation 3 below,

[0397] (Equation 3)

[0398] where S is a particular sample quality metric calculated as a function of N peak metrics, x1, x2,..., x N and a i are constants.

[0399] Reference spectroscopy, statistical analysis, and machine learning

[0400] In some embodiments, the sample quality metric can be determined based on and / or using one or more reference spectra. The reference spectra can be or include one or more spectra obtained from IR absorption measurements of a reference sample.

[0401] The reference spectra can be pre-measured and stored in, for example, a proprietary database, accessed from a public database, or measured in parallel with various processing steps, such as substantially simultaneously.

[0402] In certain embodiments, reference samples from which reference spectra can be obtained can include, but are not limited to, samples with known one or more specific sample quality metrics (e.g., determined by other orthogonal, e.g., more expensive and / or time-consuming methods, not suitable for real-time and / or on-line analysis), having known and / or desired purity, having known individual components, etc. The reference samples can be prepared to match samples or constructs screened for better stability, expression, or binding properties to their cognate substrates.

[0403] For example, in certain embodiments, one or more reference samples with known purity of, for example, a specific molecular species (e.g., protein, nucleic acid, virus particle) and / or its form (e.g., specific secondary structure, glycosylation, monomer purity) can be obtained and their IR spectra measured. In certain embodiments, the reference samples can be (e.g., deliberately) doped with one or more impurities such as waste, undesired molecular forms, (e.g., known) biological process inputs that may not be completely converted and / or filtered out by upstream processing, etc., and the corresponding IR spectra measured.

[0404] In certain embodiments, reference spectra can be obtained and / or modified in computer simulations through various computational processes. For example, reference spectra can be constructed by ab initio methods for specific molecular structures. Initial (e.g., library) reference spectra can be combined, scaled, or otherwise preprocessed, e.g., according to Beer's law, to create new customized reference spectra that, for example, capture specific variations in sample parameters of interest, remove baselines, reflect sub-band deconvolution, display second derivative spectra, etc., and are thus customized for specific sample quality metrics and / or samples.

[0405] In certain embodiments, based on a comparison between a specific target spectrum measured from a sample with unknown (e.g., completely or partially unknown) characteristics and one or more such reference spectra, one or more sample quality metrics reflecting the sample purity, for example, can be determined.

[0406] For example, in certain embodiments, the reference spectrum methods described herein can be applied to the measurement of gene therapy products (such as virus vector samples). For example, one or more reference spectra can be obtained from high-quality reference samples, for example, including desired purity in terms of the fraction of full capsids (e.g., fraction of full capsids above a specific threshold), monomeric particles, lack of certain impurities (e.g., host cell proteins and / or host cell nucleic acids; e.g., aggregates; e.g., fragments), etc. In certain embodiments, for example, a high-quality virus vector sample can have a full capsid fraction above a specific threshold (such as 70%, 80%, 90%, etc.).

[0407] An aqueous sample comprising one or more viral vector species can be interrogated, for example, subsequently or in parallel, by the mid-infrared analyzers described herein to obtain one or more target IR absorption spectra. The one or more target IR absorption spectra can then be compared to a high-quality reference spectrum to determine a measure of the sample quality.

[0408] For example, Fig.12 Illustrative absorbance plot 1200 shows an exemplary (scaled) high-quality reference spectrum 1202 of a high-quality viral vector sample having a high content of full capsids (e.g., about or better than 80% full capsids) compared to spectrum 1204 of a lower-quality viral vector sample having a low content of full capsids. The difference between the two spectra can be observed in absorbance plot 1200. In certain embodiments, a difference spectrum can be determined by subtracting the target spectrum from the high-quality reference spectrum (or vice versa) to show the variation in absorbance as a function of wave number. For example, Fig.12 Illustrative difference spectrum 1220 determined by subtracting high-quality reference spectrum 1202 from low-quality viral vector sample spectrum 1204 is shown. In difference spectrum 1220, effects such as a frequency shift (especially a red shift (e.g., a shift to lower frequencies)) in the amide-II band and a reduction in absorption in the amide-III region are evident as an asymmetric line shape 1222 and a pair of negative peaks 1224, respectively. One or both of these features and / or various peak metrics calculated therefrom can be used to determine a sample quality metric indicative of various properties of the viral vector sample quality.

[0409] Comparison of the target IR absorption spectrum to one or more reference spectra can be done in various ways, such as calculating a difference spectrum to obtain a comparison spectrum. In certain embodiments, a numerical measure of similarity can be calculated using the target spectrum and the reference spectrum, a preprocessed version thereof (e.g., derivative spectrum, scaled and / or baseline-corrected spectrum, etc.) and / or a comparison spectrum calculated therefrom. The numerical similarity measure can include, but is not limited to, a correlation value, a covariance value, a Pearson correlation value, an overlap integral, etc.

[0410] In certain embodiments, one or more machine learning models can be used to determine one or more sample quality metrics from IR absorbance data. Among other things, as an example, various reference spectra can be used to train the machine learning model in order to adjust and / or optimize the variable (learnable) parameter weights in one or more network layers. Once trained, the machine learning model can then be used to infer, i.e., determine metrics from new unknown sample spectra. For example, in certain embodiments, e.g., the machine learning model can receive an IR spectrum as input and generate a determined value of one or more sample quality metrics as output (e.g., by inference). In certain embodiments, the machine learning model receives a single IR spectrum (e.g., corresponding to a single time point) as input. In certain embodiments, the machine learning model receives multiple IR spectra (e.g., collected at different time points) as input.

[0411] In certain embodiments, a similarity score can be determined, for example, based on correlation values, covariance values, Pearson correlation values, overlap integrals, etc., as well as the machine learning-based techniques described herein. In certain embodiments, the similarity score can be generated and updated substantially in real time. For example, reference spectra of full and empty capsid samples can be obtained (e.g., loaded, received, or otherwise accessed) by a mid-infrared analyzer and / or a processor in communication therewith. Since mid-infrared absorption data is obtained repeatedly over time, e.g., to monitor virus vector production, purification, etc. (e.g., for AAV samples), the similarity score can be generated in real time and displayed (e.g., to provide a real-time view of quality), stored (e.g., as a quality log), and / or provided for further processing, e.g., to control various parameters of the production unit described herein.

[0412] Data preprocessing

[0413] In certain embodiments, the IR absorbance data (such as mid-infrared spectral data) used to determine one or more sample quality metrics is preprocessed data. For example, in certain embodiments, one or more preprocessing steps can be performed on the mid-infrared spectral data before the mid-infrared spectral data is used to calculate a particular sample quality metric and / or used as input to a machine learning model.

[0414] For example, in certain embodiments, the mid-infrared spectral data can be preprocessed by a baseline correction method that removes background signals, such as background absorption from the mobile phase (e.g., the H-O-H bending mode of water), in order to obtain a mid-infrared absorption spectrum indicative of one or more analytes in the mobile phase. This method (baseline correction by removing background signals) can be referred to as background subtraction and is achieved, for example, by subtracting a reference (e.g., background absorbance) spectrum A from the measured spectrum A ref to obtain a background-corrected spectrum A bg= A - A ref . In certain embodiments, a reference spectrum can be calculated and / or selected based on a model (such as a hybrid model) to reflect the presence and / or changes in the amount of one or more background components in the mobile phase.

[0415] For example, in certain embodiments, one or more background components are or include water molecules such that the reference spectrum is selected or calculated to reflect the appropriate intensity / relative amplitude of absorption attributable to the water molecules present in the sample. In certain embodiments, the reference spectrum can be calculated to reflect, for example, the displacement of water molecules by analyte molecules as described herein. In certain embodiments, multiple reference spectra can be calculated, for example, to reflect the changes over time of certain background molecules during specific biological and / or sample processing steps.

[0416] For example, during chromatographic elution, parameters such as salt content, pH, etc. can be changed, for example, in a stepwise or gradient manner such that when measuring IR spectral data from an aqueous sample exiting the chromatography column, the desired time-varying protein absorption spectrum is superimposed on the time-varying background spectrum. The change in the background spectrum may be due to, for example, in the case of ion-exchange chromatography where the salt content changes and water molecules are displaced by salt. Thus, in certain embodiments, multiple and / or continuously scaled reference spectra are used to reflect the change in background absorption as the salt concentration increases. In certain embodiments, selecting and / or calculating a specific reference spectrum can include using one or more template reference spectra along with the values of input parameters (such as the salt concentration curve used by the chromatography column) and / or measurement sensor values measured by a sensor, such as conductivity or pH value.

[0417] For example, in certain embodiments, when obtaining mid-infrared spectral data during IEX chromatography column elution, the conductivity can be measured using a conductivity sensor. The conductivity measured by the conductivity sensor can then be used to select or calculate a specific reference spectrum that reflects a specific water molecule concentration as the salt concentration increases and thus salt displaces water molecules. Akhgar et al., “QCL-IR Spectroscopy for In-Line Monitoring of Proteins from Preparative Ion-Exchange Chromatography,” Anal. Chem. 94:5583 - 90 (2022), the content of which is incorporated herein by reference in its entirety, for example, using a conductivity sensor to perform background compensation due to a salt (NaCl) gradient.

[0418] In certain embodiments, IR spectra can be averaged, for example to improve their quality (e.g., signal-to-noise ratio). For example, to obtain an IR spectrum at a particular time point t1, multiple IR absorption spectra can be continuously acquired in the vicinity of t1, e.g., within a window t1 + δ, and averaged to produce a single signal-averaged spectrum. Thus, in certain embodiments, the IR spectrum corresponding to a particular time point is itself a function (e.g., an average) of multiple IR spectra collected about the particular time point.

[0419] Combining multiple time points and temporal changes

[0420] In certain embodiments, a sample quality metric can be determined based on the values of IR absorption at multiple (e.g., not necessarily just the current) time points. For example, the sample quality metric can measure the temporal variation or aggregate value of one or more features of an IR absorption spectrum. In certain embodiments, for example, in cases where a particular sample quality metric can be determined using a value calculated from an IR absorption spectrum corresponding to a measurement performed at a single (e.g., current) time point, a differential or time-aggregated sample quality metric can be determined using values calculated from multiple IR absorption spectra, each corresponding to a different time point. In certain embodiments, individual values of a particular single-time-point sample quality metric can be determined at multiple time points (e.g., each value corresponding to a particular time point), and then combined, for example, by calculating differences, averages, medians, variances, etc.

[0421] In certain embodiments, a differential sample quality metric is calculated as the difference between the values of a particular (e.g., other) sample quality metric at two time points, e.g., as the difference between consecutive times. In certain embodiments, a time-aggregated sample quality metric can be calculated based on a running sum, average, median, mode, variance, standard deviation, etc. within a particular time window (such as a particular number of seconds and / or a look-back window of measured values) and / or in a cumulative fashion, thereby aggregating the measured values from an initial time point to the current time point. Various differential and / or aggregated sample quality metrics can be determined, for example, based on the time difference, cumulative (over time) sum, time average, etc. between sample quality metrics (such as absorbance values at particular wave numbers, ratios between multiple absorbance values, and various peak metrics described herein).

[0422] In certain embodiments, the spectra can be manipulated so as to emphasize particular features and / or variations of interest when calculating differential and / or time-aggregated metrics. For example, in certain embodiments, normalization methods can be used, which allow the creation of normalized spectral difference metrics that facilitate the identification of compositional changes in a sample.

[0423] Turning to the following equation (4-6), the spectrum can be normalized by reference to yield a normalized spectrum of the sample that is independent of the concentration of a particular composition (e.g., a single particular analyte or a composition of one or more analytes). That is, due to a particular composition at concentration C in the sample, the absorbance at a particular wavenumber ν i is

[0424] (Equation 4) A(ν i ) = A i = ε i × C × l,

[0425] where l is the path length.

[0426] This absorbance can be normalized by dividing by the reference absorbance A i obtained at the reference wavenumber ν ref , where

[0427] (Equation 5) A ref = A(ν ref ) = ε ref × C × l

[0428] In this way, the normalized absorbance spectrum A / A ref can be determined such that at each wavenumber ν i , the normalized absorbance is:

[0429] (Equation 6)

[0430] Thus, the normalized absorbance spectrum can be determined according to Equation 6 by dividing by the reference value obtained at a particular wavenumber. For non-normalized absorption spectra acquired at multiple time points, if the composition and concentration remain constant, there will be no difference between the spectra acquired at multiple (e.g., consecutive) time points. However, a change in concentration will result in an overall increase in absorption at each wavelength such that the difference in absorption at each wavenumber will be approximately proportional to the change in concentration. If normalization according to Equation 6 is performed to obtain a normalized spectrum at each time point, then a change in concentration will not affect the spectral difference at one or more wavelengths. That is, at each wavenumber ν i , the change Δ in the normalized absorbance due to a change in concentration rather than composition is equal to zero, plus or minus (e.g., small) baseline noise term χ i

[0431] (Equation 7a)

[0432] However, if the composition changes, the normalized spectral difference (Δ) will contain an additional contribution beyond the baseline noise term (e.g., Δ = μi ±χ i )。Therefore, by detecting whether the change in the normalized spectral difference is pure noise or contains additional factors, the change in the sample composition can be determined. The composition can be a single analyte, or a mixture of different analytes and / or species, forms, etc. of the analyte. Thus, changes in composition can occur due to the addition of new different analytes, as well as due to differences in the relative fractions (e.g., ratios between) of a particular analyte and / or its species. In the presence of a mixture, a change in concentration (e.g., as opposed to composition) refers to the total concentration of the mixture, keeping the ratio between its components constant. As described above, such a change in concentration does not affect the normalized absorbance spectrum.

[0433] The normalized spectral difference Δ can be determined in various ways. For example, Equation 7a shows the difference at a particular wavelength. In some embodiments, the normalized spectral difference Δ can be calculated by taking an integral over one or more specific spectral bands, as shown in Equation 7b below. The one or more specific spectral bands can include any of the spectral bands described herein, such as amide I, amide II, amide III, asymmetric and / or symmetric PO4 stretching bands, as well as other spectral bands of interest not necessarily described herein.

[0434] (Equation 7b)

[0435] where the integral is taken over one or more specific spectral bands of interest (e.g., a single continuous region and / or a combination of two or more non - continuous spectral regions).

[0436] Accordingly, the normalized spectral difference can be determined and monitored in real time and used to identify whether and / or when a compositional change occurs. In some embodiments, identifying a compositional change can include analyzing the normalized spectral difference signal to detect the occurrence of a change point, for example using various methods including but not limited to those described in R. Killick, P. Fearnhead, and I. A. Eckley, "Optimal detection of changepoints with a linear computational cost," Journal of the American Statistical Association, Vol. 107, No. 500, 2012, pp. 1590-1598. In some embodiments, a step change can be detected to identify a compositional change. In some embodiments, a change in one or more statistical properties of the normalized spectral difference signal can be used to identify a change in composition, for example based on whether they exceed a particular threshold and / or move outside of a particular window (e.g., an acceptable range). In some embodiments, multiple changes in composition can be identified, for example, particularly due to multiple changes in the analyte composition, addition of different analytes at different times, and conformational changes induced by aggregation, temperature, and / or other buffers (e.g., salt gradients), etc. (e.g., any change in the spectral line shape). These changes in concentration can occur on different timescales and / or alter various statistical properties in unique ways and can thus be used to distinguish between various different mechanisms for changing the sample composition.

[0437] Equations 7a and 7b show subtracting the absorbance at time t + Δt from an earlier measurement of the absorbance at time t. Additionally or alternatively, it should be understood that the normalized spectral difference can be determined by subtracting in the other direction, i.e., subtracting previous data from the most recent (e.g., current) data, as shown in Equations 7c and 7d below:

[0438] (Equation 7c)

[0439] (Equation 7d)

[0440] Accordingly, the normalized spectral difference signal can be used to monitor compositional changes in any stream monitored by IR spectroscopy. That is, while a change in analyte concentration will produce a nominal zero change in the difference between the normalized absorbance spectra, a change in the sample composition will produce a change in the differential spectrum that appears above the baseline cumulative noise. This method can be used to monitor the composition of column effluents, the composition of the retentate and / or permeate in a UF / DF process, and the progress of a chemical reaction.

[0441] In certain embodiments, the IR absorbance spectrum and / or a function thereof (e.g., a pre-processed or adjusted version of the IR absorbance spectrum) can be integrated, subtracted, etc., such as those shown in Equations 7a to 7d above. For example, in certain embodiments, the IR absorbance spectrum can be a baseline-corrected spectrum. In certain embodiments, the absolute value, square, shifted version, etc. of the IR absorption spectrum can be used, such as to ensure a particular (e.g., positive) sign of the differential signal, as those shown and / or a time-aggregated signal.

[0442] B.ii. Combination with additional sensors

[0443] In certain embodiments, the process monitoring and / or control techniques described herein can incorporate and / or use data generated by one or more additional sensors (e.g., in addition to the mid-infrared analyzer). In certain embodiments, the one or more additional sensors include sensors for measuring parameters such as temperature, pressure, pH, conductivity, flow rate, etc. Such sensors can include but are not limited to one or more temperature sensors, one or more pressure sensors, one or more pH sensors, one or more conductivity sensors, one or more flow rate sensors, optical sensors, etc.

[0444] In certain embodiments, the one or more additional sensors can include sensors that are also capable of measuring one or more physical, chemical, biological, or microbial properties of one or more analytes within a sample. In certain embodiments, one or more additional sensors can be used, such as in conjunction with one or more mid-infrared analyzers, to generate data for determining one or more sample quality metrics. In certain embodiments, for example, mid-infrared spectral data can be used in conjunction with data from one or more additional sensors to determine one or more sample quality metrics.

[0445] For example, in certain embodiments, the additional sensor is or includes a UV absorption sensor. A UV absorption sensor can be or include any sensor operable to measure the absorption of a sample in the UV (e.g., from about 200 nm to about 300 nm) spectral range. In certain embodiments, the UV absorption sensor can be or include a fixed path length sensor that measures the UV absorption of a sample using a fixed path length cell. In certain embodiments, the UV absorption sensor can be or include a slope spectrum sensor, such as CTech TM and / or its variants / embodiments that measure UV absorption while changing the path length through the sample. In particular, in certain embodiments, UV absorption measurements can be used in conjunction with mid-infrared absorption measurements as, for example, an internal check and / or complementary technique. For example, in certain embodiments, the combination and / or extended range of data collection can provide intrinsic verification and quantification of different types of data, etc. For example, in certain embodiments, the total protein concentration can be measured using one or both of UV absorption and mid-infrared absorption spectral data. In certain embodiments, additionally or alternatively, labeled and tagged solutions in mid-infrared will increase the value of UV-based methods as well as for species identification, anti-counterfeiting, and product traceability.

[0446] B.iii Real-time control of PAT system

[0447] In certain embodiments, data analysis, machine learning, and artificial intelligence, etc. can be utilized in conjunction with the IR spectral measurements and / or sample quality metrics determined as described herein, not only to provide real-time monitoring and reporting of sample quality metrics, but also can be additionally or alternatively used for predictive analysis, thereby analyzing the data collected in real-time and additionally or alternatively analyzing it in conjunction with historical data sets and / or empirical models to predict the future state of the process and / or the quality of materials in the process flow.

[0448] For example, the IR measurements and data analysis tools described herein can provide detailed information about process parameters, process performance, and process stability. They can be used to control process parameters, for example, to improve performance in various growth and purification (filtration) steps, including but not limited to cell culture, virus production, clarification, concentration, diafiltration, chromatography, purification, dead-end filtration, tangential flow filtration, tangential flow depth filtration. Additionally or alternatively, predictive analysis can be utilized during analytical development, process development, and formulation to design experiments to develop a very detailed understanding of the process, potential failure mode and effect analysis, and sensitivity analysis during the modeling process.

[0449] Production Unit

[0450] Methods such as these can be used for upstream processes (such as cell culture and harvest) as well as downstream applications (including purification / filtration, formulation, and filling and finishing unit processes). For example, Fig.13 illustrates various upstream and downstream bioproduction process steps and characteristics that can be monitored particularly by the MIR analyzer 1302 and / or used to control process parameters by the systems and methods described herein.

[0451] For example, as Fig.13As shown, the raw materials used in the production of biologic products (e.g., proteins, nucleic acids, viral vectors, etc.) can be initially tested and / or monitored when they are provided to, for example, various processing steps and production units to evaluate sample quality metrics related to, among other things, material purity, identity, and the presence of impurities. The various inputs, outputs, and parts of a bioreactor production unit (such as a seed bioreactor and / or a production bioreactor) can be monitored using the techniques described herein, for example, to confirm the identity of the desired biomolecule produced, determine the titer, and ensure the presence and / or provision of sufficient nutrients in the bioreactor.

[0452] In certain embodiments, for example, one or more of the MIR analyzers described herein can be used in conjunction with purification (filtration) units such as alternating tangential flow filtration (ATF) systems, tangential flow depth filtration (TFDF) systems, tangential flow filtration (TFF) systems, chromatography columns, direct or conventional flow filtration, ultrafiltration, diafiltration, etc.

[0453] In certain embodiments, for example, one or more sample quality metrics and / or IR spectral data can be provided to control software and / or hardware systems and components such as supervisory control and data acquisition (SCADA), manufacturing execution system (MES) systems, etc. for making decisions regarding changes to process control parameters and product release.

[0454] For example, when processing proteins, data such as the empty / full ratio (e.g., in the context of viral vector production processes) or the presence of high molecular weight species (such as dimers, trimers, and multimers) can be used to adjust process parameters such as flow rate, flow direction, pressure, temperature, pH, etc. In certain embodiments, parameters such as the addition and / or variation in the amount of raw materials in the process stream can be adjusted. Process parameters controlled and / or adjusted as described herein can include, but are not limited to, process parameters that affect certain attributes that should be within appropriate limits, ranges, or distributions to ensure the desired product quality (i.e., CQA) (e.g., process parameters that affect CQA, referred to as "critical process parameters (CPP)").

[0455] Additionally or alternatively, data characterizing protein content, secondary structure, protein aggregation, viral vector content, empty-full ratio, capsid aggregation, etc. in real time (such as specific sample quality metrics) can be used to guide decisions such as when to stop processing, open or close valves, direct sample collection and fraction collection, and combine different materials. For example, in certain embodiments, a decision regarding column loading can be made by monitoring the breakthrough of a chromatography column. For example, in one instance of control, upon detection of breakthrough, a command can be sent to a continuous production system to open / close valves, redirect the process flow to the next inline chromatography column, and / or transition a full column to the elution and / or wash phase of the process.

[0456] The methods described herein can be used with a variety of chromatography columns, including but not limited to affinity chromatography (AC) (e.g., Protein A) columns, hydrophobic interaction chromatography (HIC) columns, ion exchange chromatography (IEX) columns, size exclusion chromatography (SEC) columns, and mixed-mode chromatography columns [e.g., those implementing any combination of the foregoing chromatography columns (e.g., IEX and HIC; e.g., IEX and SEC)].

[0457] In certain embodiments, the methods described herein can be used in conjunction with a TFF concentration production unit, e.g., during a viral vector and / or antibody production process, to monitor aggregation and continue and / or stop concentration, e.g., based on a measured aggregation level.

[0458] For example, in certain embodiments, during the discovery and manufacture of protein therapeutics and / or diagnostic reagents (including, e.g., monoclonal antibodies, fusion proteins, viral capsid proteins, antibody-drug conjugates, etc.), it is highly desirable to accurately measure the total concentration (also referred to as titer) of an aqueous protein mixture and the degree of their molecular heterogeneity over a wide dynamic concentration range (e.g., from about 1 picogram / liter to about 500 grams / liter). Additionally or alternatively, it is desirable to perform such measurements under static or dynamic flow conditions of up to 100 liters / minute or higher.

[0459] Protein heterogeneity and chromatographic elution control

[0460] In certain embodiments, the sample mass metric calculated from the IR absorption spectra described herein can be used to measure different types of molecular heterogeneity. For example, in certain embodiments, a mixture may be heterogeneous due to having different species of free (unbound) proteins. In certain embodiments, the heterogeneity can be produced by aggregation, where the mixture includes populations of the same protein, each protein being free (unbound) or chemically bound to one (dimer) or more (trimer, tetramer, pentamer, etc.) of the same protein, thus forming different degrees of aggregation. Additionally or alternatively, the desired form of the molecular heterogeneity to be measured can include populations of the same protein having different types or degrees of molecular conjugation (e.g., polysaccharides, glycans, polyethylene glycol, or small molecules for therapeutic means).

[0461] In certain embodiments, the methods described herein can quantify protein heterogeneity, and can be quantified in a variety of different ways. For example, in certain embodiments (e.g., as further described in detail below in certain embodiments), a protein aggregation metric can be determined that quantifies the total aggregation level or percentage of the aggregation level. In certain embodiments, a conjugation-based metric can be calculated from the IR spectrum that quantifies the total level or percentage level of glycosylation or small molecule conjugation. In certain embodiments, the molecular weight can be determined and a histogram of the molecular weight can be shown.

[0462] In some embodiments, the level of quantitative aggregation is particularly important because aggregates can reduce the efficacy and safety of a pharmaceutical product.

[0463] For example, during ion exchange chromatography (IEX), a heterogeneous aqueous protein mixture is deliberately and timely separated into its individual protein components over time by using a chromatography column and an ionic salt gradient. In some embodiments, it is desirable to continuously and non-invasively monitor the total protein concentration and the degree of heterogeneity of the eluate in real time such that the collection window for the target protein can be dynamically controlled to produce optimal results (e.g., the highest monomer purity). In other words, it may be necessary to accurately quantify the level of protein aggregation (high molecular weight species) that most often occurs during the elution tail. Optimizing the collection window can maximize the protein target quality and yield while also maximizing the lifetime of the column.

[0464] In some embodiments, columns are increasingly loaded with higher concentrations of total protein to maximize the target protein yield and extend the useful life of the column. However, such practices tend to result in the formation of more aggregates and a reduction in the separation time between the target protein (monomer) peak and the first protein aggregate peak (dimer). That is, the monomer and dimer peaks become increasingly overlapped and thus less distinct, posing a challenge to determining the appropriate (e.g., optimal) target protein capture window.

[0465] In some embodiments, the methods described herein can be used for other types of separations (e.g., not limited to aggregate removal) where product (e.g., or impurity) breakthrough may occur, such as during clarification prior to chromatography to remove HCP and / or DNA onto a flow-through filter in flow-through mode.

[0466] In some embodiments, the systems and methods described herein can be used to monitor protein secondary / tertiary / quaternary structure. Secondary / tertiary / quaternary structure metrics can be used to control processes in antibody production as well as in viral vector sample production (e.g., by monitoring capsid protein secondary / tertiary / quaternary structure). In one example, changes in the secondary / tertiary / quaternary structure of a protein molecule can be detected and used, for example, to divert flow in different ways as an action (e.g., based on an electronic trigger signal).

[0467] Sample quality metrics indicating protein aggregation, secondary structure motifs, and other properties can be monitored, for example, by calculating one or more peak metrics from an IR absorption spectrum as described in the following examples. For example, in certain embodiments, the frequency positions of one or more of the amide-I band, amide-II band, and amide-III band can be determined, such as the center frequency and / or centroid frequency described herein, and monitored to track the level of protein aggregation. In certain embodiments, the sample quality metric is a protein aggregation metric indicating the level of protein aggregation within the sample. In certain embodiments, the protein aggregation metric is determined based on (e.g., as) the frequency position (e.g., center frequency; e.g., centroid frequency) of the amide I band. In certain embodiments, the protein aggregation metric is determined based on (e.g., as) the frequency position (e.g., center frequency; e.g., centroid frequency) of the amide II band. In certain embodiments, the protein aggregation metric is determined based on (e.g., as) the frequency position (e.g., center frequency; e.g., centroid frequency) of the amide III band.

[0468] Viral vector production monitoring and control

[0469] In certain embodiments, the methods described herein can be used to monitor and control viral vector production control. In certain embodiments, the techniques described herein are applicable to various viral vector production methods, including, for example, transfection-based techniques and techniques utilizing stable production cell lines (e.g., for producing viral vectors or other products, such as antibodies). Among them, viral vectors are a class of macromolecules that, in certain embodiments, have a molecular weight greater than 1 MDa. Viral vectors can be used to infect targeted host cells with genetic material, for example, for purposes such as editing the genome of the infected cells or directly translating specific proteins within the infected cells. Thus, viral vectors for editing the genome of infected cells can be used as gene therapy devices. In certain embodiments, viral vectors can be used to trigger an immune response, for example, to achieve a vaccine function, by directly translating specific proteins within the infected cells.

[0470] Viral vectors include, but are not limited to, adenoviruses, adeno-associated viruses (AAVs), retroviruses (e.g., lentiviruses), and plant-based viruses (e.g., tobacco mosaic virus). As Fig.14 shown, the viral vector capsid includes (e.g., loaded with) a gene cassette. During production, certain viral vector particles can be empty, lacking the required genetic payload to be delivered to the infected cells. Thus, in certain embodiments, determining not only the total amount of viral particles but also the fraction of full viral particles that encapsulate the required genetic payload and are thus suitable for their intended application is valuable for controlling production and evaluating process yields.

[0471] Structurally, the viral vector includes a (e.g., approximately spherical) protein capsid with a diameter of about 100 nm or less and includes (e.g., encapsulates) one or more nucleic acid strands, such as DNA or RNA (e.g., mRNA) of different lengths in terms of the number of nucleotide bases. Thus, as described herein, methods for quantifying the protein and nucleic acid content in a mixture can be used to determine sample quality metrics such as total nucleic acid content (e.g., concentration), viral capsid content (e.g., concentration), and the fraction of full capsids. In particular, as described herein, the mid-infrared organic fingerprint bands spanning from approximately one thousand (1,000) to one thousand eight hundred (1,800) wavenumbers include multiple spectral sub-regions (sub-bands) that can be assigned to the capsid, the genetically active payload (cargo) contained within the capsid, or a combination of the capsid and the genetically active payload. The methods described herein can utilize the entire range of the mid-infrared fingerprint spectral region (from about 1,000 cm -1 to about 1800 cm -1 -1) and / or its various sub-bands. Such methods can quantify the concentration of the capsid, the genetic material, and / or their volume ratio, and additionally or alternatively, quantify the difference between the composite spectrum of the sample and the composite spectrum of a purified high-quality sample.

[0472] For example, as described and Fig.11 shown herein, various spectral bands can be used to measure the protein and / or nucleic acid content in a sample (including in a mixture), where certain bands (such as amide-II and / or amide-III) can be used to independently quantify the protein content relative to the nucleic acid concentration. Since the viral vector is structurally a protein capsid encapsulating nucleic acid material, spectral absorption within these bands (e.g., together with the Beer Lambert law) can be used to determine protein content metrics and nucleic acid metrics that separately quantify the protein and nucleic acid concentrations, which in turn can be used to quantify the (protein) capsid and (nucleic acid) payload metrics.

[0473] For example, in certain embodiments, the total capsid concentration can be calculated based on a protein content metric (e.g., a scaled version of the protein content metric), which can be determined based on absorption in the amide-I, amide-II, and / or amide-III regions. In certain embodiments, use of the amide-II and / or amide-III spectral regions is desirable and aids in the independent quantification of protein content, as the nucleic acid spectra have relatively little absorption in those (amide-II and amide-III) regions. The absorption intensity in the amide-II and / or amide-III regions can be determined using a peak intensity metric that measures the intensity of the amide-II and / or amide-III band, such as peak amplitude or area under the curve (AUC) measurement. Thus, the protein content metric and / or the total capsid content can be determined based on the peak intensity measures of one or both of the amide-II and amide-III bands (e.g., a linear combination).

[0474] In certain embodiments, the nucleic acid content metric, such as the total nucleic acid concentration, can be calculated by one or more peak metrics calculated based on the asymmetric and / or symmetric PO4 bands. The absorption intensity in the asymmetric and / or symmetric PO4 regions can be determined using a peak intensity metric that measures the intensity of the asymmetric PO4 and / or symmetric PO4 band, such as peak amplitude (“peak”) or area under the curve (“AUC”) measurement. Thus, the nucleic acid metric can be determined based on the peak intensity measurements of one or both of the asymmetric PO4 and / or symmetric PO4 bands (e.g., a linear combination).

[0475] In certain embodiments, the protein content metric and / or the nucleic acid metric can be used to calculate the full capsid fraction, which provides a measure of the fraction (e.g., ratio, percentage, etc.) of full (i.e., successfully loaded with the desired genetic payload) capsids. For example, the full capsid fraction can be determined based on the ratio of (i) the nucleic acid content metric to (ii) the protein content metric and / or the total capsid content determined therefrom.

[0476] In certain embodiments, other peak metrics can be used to determine sample quality metrics, such as capsid content and / or capsid integrity fraction. For example, in certain embodiments, the frequency positions (e.g., center frequency, centroid frequency) of one or more of the amide-I band, amide-II band, amide-III band, asymmetric PO4 band, and symmetric PO4 band can be determined. In certain embodiments, the frequency positions can indicate a specific percentage content of full capsids versus empty capsids.

[0477] In some embodiments, additionally or alternatively, one or more peak metrics can be used to determine the aggregation level of capsids in a sample, e.g., based on frequency position changes. In some embodiments, as described herein, one or more protein secondary structure metrics of a viral vector sample can be determined. For example, combining ion exchange chromatography (charge separation of viral capsids) with mid-infrared measurements (changes in protein secondary structure) can result in higher purity by selectively separating intact capsids.

[0478] Sample quality metrics related to the production of viral vectors described herein can be stored, displayed, or provided, e.g., as trigger signals, for monitoring, interactively adjusting, and / or automatically tuning process parameters. Fig.15 An exemplary process 1500 for real-time evaluation and monitoring of viral process production by mid-infrared spectroscopy is shown. In some embodiments, an original IR absorption spectrum 4302 of an aqueous sample comprising a viral vector species is collected. In some embodiments, various preprocessing steps, such as smoothing, decimation, baseline correction (e.g., to account for water temperature drift, water displacement, etc.), etc., can be performed 1504. The IR absorption spectrum can then be used to determine sample quality metrics, such as protein content, nucleic acid content, viral capsid content, and intact capsid fraction 1506. In some embodiments, sample quality metrics such as these can be CQAs, and / or used to determine one or more CQAs and / or CPPs. In some embodiments, the sample quality metrics and / or parameters determined therefrom can be displayed 1508a and / or stored in a memory 1508c. In some embodiments, the determined sample quality metrics, such as viral capsid content and / or intact capsid fraction, can be used to generate digital and / or analog electronic trigger signals 1508b, which can be used, e.g., for feedforward and / or feedback process control. For example, processes such as flow rate, flow direction, pressure, temperature, pH, etc., can be adjusted as described herein. In some embodiments, the trigger signal can be used to guide decision-making, such as when to stop processing, open or close valves to direct sample collection and fraction collection, e.g., to collect specific fractions of a sample eluted from a chromatography column, which include, e.g., high-titer and / or high-intact capsid fractions.

[0479] Go to Fig.16A and 16B such trigger signals can be used for various processing steps in viral vector production. Fig.16A and 16B show process flows for AAV and lentiviral vector production, respectively. Thus, viral capsid content and / or full fraction can be monitored from aqueous samples, e.g., during steps and / or individual steps including production and subsequent steps.

[0480] For example, on-line measurements of the virus vector full fractions (e.g., for AAV capsids, lentiviral capsids, etc.) can be used to monitor the separation of empty capsids and full capsids during chromatographic purification steps. Chromatographic purification can, for example, separate empty capsids from full capsids using anion exchange (AEX) chromatography. In certain embodiments, other chromatographic techniques can be used, such as affinity chromatography (AC) columns, hydrophobic interaction chromatography (HIC) columns, ion exchange chromatography (IEX) columns, size exclusion chromatography (SEC) columns, and mixed-mode chromatography columns [e.g., which implement any combination of the foregoing columns (e.g., IEX and HIC; e.g., IEX and SEC)]. In certain embodiments, on-line monitoring of the aqueous sample as it elutes from the chromatography column can allow for observations and / or control based on the measured virus capsid content and / or full fraction to select specific target fractions of the eluate to retain or discard according to the desired target purity and / or yield.

[0481] Go to Fig.17 , in certain embodiments, the reference spectra described herein can be used to determine sample quality metrics. For example, as Fig.17 shown, in an exemplary process 1700, the raw spectral data can be converted into a quantitative sample quality metric, such as a CQA, and then displayed, stored, or used to trigger process control. As Fig.17As shown, the original IR absorption spectrum 1702 of an aqueous sample including a viral vector species is collected. In certain embodiments, various preprocessing steps can be performed, such as smoothing, extraction, baseline correction (e.g., to account for water temperature drift, water displacement, etc.) 1704. In certain embodiments, the original spectrum 1702, with or without preprocessing 1704, can be scaled 1706, e.g., to adjust the overall amplitude. In certain embodiments, the original spectrum can be scaled by a constant equal to and / or based on (e.g., determined using) one or more peak metrics (such as peak amplitude, area under the curve, etc.) calculated for one or more specific absorption bands. As described herein, the target spectrum can be compared to one or more reference spectra (such as a high-quality viral vector spectrum) to determine one or more sample quality metrics 1708. The comparison between the target spectrum and the reference spectrum can include calculating a difference spectrum, a differential first derivative, a differential second derivative, correlation, covariance, Pearson correlation, overlap integral, etc. In certain embodiments, the comparison between the target spectrum and the reference spectrum results in a comparison spectrum, such as a difference spectrum, which can be used to calculate peak metrics and ultimately calculate the sample quality metrics. In certain embodiments, the sample quality metrics can be calculated based on the target spectrum and the reference spectrum without having to calculate the comparison spectrum, e.g., by calculating covariance, overlap integral, etc. In certain embodiments, sample quality metrics such as these can be CQAs, and / or used to determine one or more CQAs and / or CPPs. In certain embodiments, the sample quality metrics and / or parameters determined therefrom can be displayed 1710a and / or stored in a memory 1710c. In certain embodiments, the determined sample quality metrics, such as viral capsid content and / or full capsid fraction, can be used to generate digital and / or analog electronic trigger signals 1710b, which can be used for, e.g., feedforward and / or feedback process control. For example, as described herein, processes such as flow rate, flow direction, pressure, temperature, pH, etc. can be adjusted. In certain embodiments, the trigger signal can be used to guide decision-making, such as when to stop processing, open or close valves to direct sample collection and fraction collection, e.g., to collect specific fractions of the sample eluted from a chromatography column, which include, e.g., high-titer and / or high full capsid fractions.

[0482] Integrate MIR analyzers into production cells and / or process flows

[0483] In certain embodiments, one or more mid-infrared analyzers are incorporated into the production unit and / or process stream, e.g., to measure liquid inputs and / or outputs from the production unit. In certain embodiments, one or more mid-infrared analyzers are incorporated inline, e.g., by shunting, whereby a portion of the input / output stream is sampled by the analyzer and then returned to the main process stream for continued processing.

[0484] In certain embodiments, the ATR sampling geometry provides advantages for integration with pilot and / or commercial scale manufacturing production units that rely on flow in / out through large diameter channels having diameters ranging from one millimeter or greater to 1 / 4 inch to one inch. In certain embodiments, transport over 40 microns (e.g., over 100 microns) may be impracticable and thus the ability of the ATR geometry to provide a fixed, limited path length via an evanescent wave significantly facilitates integration with production scale systems, independent of the size of the channel through which the liquid flows.

[0485] B.iv Control Signal Implementation

[0486] In certain embodiments, the processing, communication, instrument control, etc. described herein may be implemented in whole or in part by various components associated with a production unit, a central processing system, or a remote device. For example, various processing, communication, and control steps may be performed by one or more of firmware embedded on one or more particular devices (e.g., mid-infrared analyzers, other complementary sensor systems, production unit controllers, etc.), a connected microprocessor system, and / or software in an external computer (e.g., directly connected, connected via Ethernet, or cloud-based). In certain embodiments, the connected computer may directly control process parameters or may transmit and / or data, process commands, control signals, etc. via one or more communication channels.

[0487] A variety of communication channels may be used, including but not limited to data packets shared via a communication network and / or communication port, analog signals transmitted as voltage or current to a device (such as a PLC or another computer) that can decode the signal, standardized communication protocols / systems (such as OPC-UA, Profibus, ModBus, etc.), or a proprietary communication channel. The communication channel may be wired or use a wireless protocol such as Bluetooth, WiFI, RF, or other technologies. In certain embodiments, the data and / or commands are encoded prior to transmission and / or sharing and are decrypted and used by the receiving device (e.g., to adjust process parameters).

[0488] C. Computer system and network environment

[0489] Turning Fig.18 , an implementation of a network environment 1800 for providing the systems, methods, and architectures described herein is shown and described. Briefly summarized, now refer to Fig.18, a block diagram of an exemplary cloud computing environment 1800 is shown and described. The cloud computing environment 1800 may include one or more resource providers 1802a, 1802b, 1802c (collectively 1802). Each resource provider 1802 may include computing resources. In some embodiments, the computing resources may include any hardware and / or software for processing data. For example, the computing resources may include hardware and / or software capable of executing algorithms, computer programs, and / or computer applications. In some embodiments, exemplary computing resources may include application servers and / or databases having storage and retrieval capabilities. Each resource provider 1802 may be connected to any other resource provider 1802 in the cloud computing environment 1800. In some embodiments, the resource providers 1802 may be connected via a computer network 1808. Each resource provider 1802 may be connected via the computer network 1808 to one or more computing devices 1804a, 1804b, 1804c (collectively 1804).

[0490] The cloud computing environment 1800 may include a resource manager 1806. The resource manager 1806 may be connected to the resource providers 1802 and the computing devices 1804 via the computer network 1808. In some embodiments, the resource manager 1806 may facilitate the provision of computing resources from one or more resource providers 1802 to one or more computing devices 1804. The resource manager 1806 may receive requests for computing resources from a particular computing device 1804. The resource manager 1806 may identify one or more resource providers 1802 capable of providing the computing resources requested by the computing device 1804. The resource manager 1806 may select a resource provider 1802 to provide the computing resources. The resource manager 1806 may facilitate the connection between the resource provider 1802 and the particular computing device 1804. In some embodiments, the resource manager 1806 may establish a connection between a particular resource provider 1802 and a particular computing device 1804. In some embodiments, the resource manager 1806 may redirect a particular computing device 1804 to a particular resource provider 1802 having the requested computing resources.

[0491] Fig.19 Examples of a computing device 1900 and a mobile computing device 1950 that may be used to implement the techniques described in the present disclosure are shown. The computing device 1900 is intended to represent various forms of digital computers such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframes, and other suitable computers. The mobile computing device 1950 is intended to represent various forms of mobile devices such as personal digital assistants, cellular telephones, smart phones, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are only intended to be examples and are not intended to be limiting.

[0492] The computing device 1900 includes a processor 1902, a memory 1904, a storage device 1906, a high-speed interface 1908 connected to the memory 1904 and a plurality of high-speed expansion ports 1910, and a low-speed interface 1912 connected to a low-speed expansion port 1914 and the storage device 1906. Each of the processor 1902, the memory 1904, the storage device 1906, the high-speed interface 1908, the high-speed expansion ports 1910, and the low-speed interface 1912 is interconnected using various buses and can be mounted on a common motherboard or in other suitable manners. The processor 1902 can process instructions for execution within the computing device 1900, including instructions stored in the memory 1904 or on the storage device 1906, to display graphical information of a GUI on an external input / output device such as a display 1916 coupled to the high-speed interface 1908. In other embodiments, multiple processors and / or multiple buses and multiple memories and various types of memories can be appropriately used. Additionally, multiple computing devices can be connected, where each device provides a portion of the necessary operations (e.g., as a server group, a set of blade servers, or a multi-processor system). Thus, as used herein, the term where multiple functions are described as being performed by a "processor" encompasses embodiments where the multiple functions are performed by any number of processors (one or more) of any number of computing devices (one or more). Additionally, in cases where a function is described as being performed by a "processor", this encompasses embodiments where the function is performed by any number of processors (one or more) of any number of computing devices (one or more) (e.g., in a distributed computing system).

[0493] The memory 1904 stores information within the computing device 1900. In some embodiments, the memory 1904 is one or more volatile memory units. In some embodiments, the memory 1904 is one or more non-volatile memory units. The memory 1904 can also be another form of computer-readable medium, such as a magnetic disk or an optical disk.

[0494] The storage device 1906 is capable of providing large-capacity storage for the computing device 1900. In some embodiments, the storage device 1906 can be or include a computer-readable medium, such as a floppy disk device, a hard disk device, an optical disk device, or a tape device, a flash memory, or other similar solid-state memory devices, or an array of devices, including devices in a storage area network or other configurations. Instructions can be stored in the information carrier. When executed by one or more processing devices (e.g., the processor 1902), the instructions perform one or more methods, such as those described above. The instructions can also be stored by one or more storage devices, such as a computer or machine-readable medium (e.g., the memory 1904, the storage device 1906, or the memory on the processor 1902).

[0495] The high-speed interface 1908 manages the bandwidth-intensive operations of the computing device 1900, while the low-speed interface 1912 manages the less bandwidth-intensive operations. This functional allocation is merely an example. In some embodiments, the high-speed interface 1908 is coupled to the memory 1904, the display 1916 (e.g., via a graphics processor or accelerator), and is coupled to a high-speed expansion port 1910, which can accept various expansion cards (not shown). In an embodiment, the low-speed interface 1912 is coupled to the storage device 1906 and the low-speed expansion port 1914. The low-speed expansion port 1914, which can include various communication ports (e.g., USB, Ethernet, wireless Ethernet), can be coupled to one or more input / output devices, such as a keyboard, a pointing device, a scanner, or a networking device such as a switch or router, for example, via a network adapter.

[0496] The computing device 1900 can be implemented in many different forms, as shown. For example, it can be implemented as a standard server 1920, or implemented multiple times in a group of such servers. Additionally, it can be implemented in a personal computer such as a laptop computer 1922. It can also be implemented as part of a rack server system 1924. Alternatively, components from the computing device 1900 can be combined with other components in a mobile device (not shown), such as a mobile computing device 1950. Each of these devices can contain one or more of the computing device 1900 and the mobile computing device 1950, and the entire system can be composed of multiple computing devices that communicate with each other.

[0497] The mobile computing device 1950 includes a processor 1952, a memory 1964, an input / output device such as a display 1954, a communication interface 1966, and a transceiver 1968, as well as other components. The mobile computing device 1950 can also be equipped with a storage device, such as a microdrive or other device, to provide additional storage. Each of the processor 1952, the memory 1964, the display 1954, the communication interface 1966, and the transceiver 1968 is interconnected using various buses, and several components can be mounted on a common motherboard or in other suitable ways.

[0498] The processor 1952 can execute instructions within the mobile computing device 1950, including the instructions stored in the memory 1964. The processor 1952 can be implemented as a chipset that includes separate and multiple analog and digital processors. The processor 1952 can provide, for example, the coordination of the other components of the mobile computing device 1950, such as the control of the user interface, the applications run by the mobile computing device 1950, and the wireless communication performed by the mobile computing device 1950.

[0499] The processor 1952 can communicate with a user through a control interface 1958 and a display interface 1956 coupled to a display 1954. The display 1954 can be, for example, a TFT (Thin Film Transistor Liquid Crystal Display) display or an OLED (Organic Light Emitting Diode) display or other suitable display technology. The display interface 1956 can include appropriate circuitry for driving the display 1954 to present graphics and other information to the user. The control interface 1958 can receive commands from the user and convert them for submission to the processor 1952. Additionally, an external interface 1962 can provide communication with the processor 1952 to enable near area communication of the mobile computing device 1950 with other devices. The external interface 1962 can provide, for example, wired communication in some embodiments, or wireless communication in other embodiments, and can also use multiple interfaces.

[0500] The memory 1964 stores information within the mobile computing device 1950. The memory 1964 can be implemented as one or more computer-readable media, one or more volatile memory units, or one or more non-volatile memory units. An extended memory 1974 can also be provided and connected to the mobile computing device 1950 through an extended interface 1972, which can include, for example, a SIMM (Single In-line Memory Module) card interface. The extended memory 1974 can provide additional storage space for the mobile computing device 1950, or can also store application programs or other information for the mobile computing device 1950. Specifically, the extended memory 1974 can contain instructions for executing or supplementing the above-described processes, and can also contain security information. Thus, for example, the extended memory 1974 can be provided as a security module for the mobile computing device 1950 and can be programmed with instructions that allow for the secure use of the mobile computing device 1950. In addition, security applications and additional information can be provided through the SIMM card, such as placing authentication information on the SIMM card in a non-hackable manner.

[0501] As described below, the memory can include, for example, flash memory and / or NVRAM memory (Non-Volatile Random Access Memory). In some embodiments, the instructions are stored in an information carrier. When executed by one or more processing devices (e.g., the processor 1952), the instructions perform one or more methods, such as those described above. The instructions can also be stored by one or more storage devices, such as one or more computer or machine-readable media (e.g., the memory 1964, the extended memory 1974, or the memory on the processor 1952). In some embodiments, the instructions can be received in a propagated signal, for example, through a transceiver 1968 or the external interface 1962.

[0502] The mobile computing device 1950 can communicate wirelessly via a communication interface 1966, which may include digital signal processing circuitry when necessary. The communication interface 1966 can provide communication in various modes or protocols, such as GSM voice calls (Global System for Mobile Communications), SMS (Short Message Service), EMS (Enhanced Message Service), or MMS messages (Multimedia Message Service), CDMA (Code Division Multiple Access), TDMA (Time Division Multiple Access), PDC (Personal Digital Cellular), WCDMA (Wideband Code Division Multiple Access), CDMA2000, or GPRS (General Packet Radio Service), etc. Such communication can occur, for example, using radio frequency via a transceiver 1968. Additionally, short-range communication can occur, such as using Wi-Fi TM or other such transceivers (not shown). Additionally, a GPS (Global Positioning System) receiver module 1970 can provide additional navigation and location-related wireless data to the mobile computing device 1950, which can be appropriately used by application programs running on the mobile computing device 1950.

[0503] The mobile computing device 1950 can also communicate audibly using an audio codec 1960, which can receive verbal information from a user and convert it into usable digital information. The audio codec 1960 can likewise generate audible sounds for the user, such as via a speaker, e.g., in the earpiece of the mobile computing device 1950. Such sounds can include sounds from a voice telephone call, can include recorded sounds (e.g., voice messages, music files, etc.), and can also include sounds generated by application programs operating on the mobile computing device 1950.

[0504] The mobile computing device 1950 can be implemented in many different forms, as shown in the figure. For example, it can be implemented as a cellular phone 1980. It can also be implemented as part of a smart phone 1982, a personal digital assistant, or other similar mobile devices.

[0505] Various embodiments of the systems and techniques described herein can be implemented in digital electronic circuitry, integrated circuits, specially designed ASICs (Application Specific Integrated Circuits), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include embodiments in one or more computer programs executable and / or interpretable on a programmable system including at least one programmable processor, which can be special purpose or general purpose, coupled to receive data and instructions from, and to send data and instructions to, a storage system, at least one input device, and at least one output device.

[0506] Actions associated with an implementation system can be performed by one or more programmable processors executing one or more computer programs. All or part of the system can be implemented as dedicated logic circuitry, such as a field programmable gate array (FPGA) or an application specific integrated circuit (ASIC), or both. All or part of the system can also be implemented as dedicated logic circuitry, such as a specially designed (or configured) central processing unit (CPU), a conventional central processing unit (CPU), a graphics processing unit (GPU), and / or a tensor processing unit (TPU).

[0507] These computer programs (also referred to as programs, software, software applications, or code) include machine instructions for a programmable processor and can be implemented in a high-level procedural and / or object-oriented programming language and / or assembly / machine language. As used herein, the terms machine-readable medium and computer-readable medium refer to any computer program product, apparatus, and / or device (e.g., a disk, an optical disk, a memory, a programmable logic device (PLD)) for providing machine instructions and / or data to a programmable processor, including a machine-readable medium that receives machine instructions as a machine-readable signal. The term machine-readable signal refers to any signal for providing machine instructions and / or data to a programmable processor.

[0508] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer. Other kinds of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form, including acoustic, speech, or tactile input.

[0509] The systems and techniques described herein can be implemented in a computing system that includes a backend component (e.g., as a data server), or includes a middleware component (e.g., an application server), or includes a frontend component (e.g., a client computer having a graphical user interface or a web browser through which the user can interact with an implementation of the systems and techniques described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), and the Internet.

[0510] A computing system may include a client and a server. The client and the server are typically far apart from each other and usually interact via a communication network. The relationship between the client and the server arises by means of computer programs that run on corresponding computers and have a client-server relationship with each other.

[0511] In some embodiments, the modules described herein may be separated, combined, or incorporated into single or combined modules. The modules depicted in the drawings are not intended to limit the systems described herein to the software architectures shown therein.

[0512] Elements of different embodiments described herein may be combined to form other embodiments not specifically set forth above. Elements may be omitted from the processes, computer programs, databases, etc. described herein without adversely affecting their operation. Additionally, the logical flows depicted in the drawings do not require the particular order or sequential order shown to achieve the desired result. Various individual elements may be combined into one or more individual elements to perform the functions described herein.

[0513] E. Examples

[0514] Example 1: Linearity at wide dynamic range and high concentration.

[0515] This example demonstrates the ability of a QCL-based IR spectrometer according to certain embodiments described herein to measure protein concentration in aqueous solutions within a wide dynamic range and with high linearity. Fig. 20A -D shows measurements of IR spectra from several different protein mixtures, each protein mixture including a specific concentration of bovine serum albumin (BSA), and a plot comparing the concentration / related peak metric measured by QCL with the a priori known concentration. Fig. 20A Several IR absorbance spectra measured for samples with different BSA concentrations are shown. Fig. 20B The values of the measured concentration are shown, which are determined by calculating the AUC of the protein absorption region by integrating each spectrum from the minimum frequency (about 1360 cm -1 ) to about 1700 cm -1 . The measured concentration is determined by scaling the determined AUC and is shown relative to the nominal reference concentration in Fig. 20B . Fig. 20C And 20D show similar measurements, where Fig.20D shows the AUC directly relative to the nominal BSA concentration, and shows the parameters of the linear fit and the R 2 value, indicating a highly linear correlation. Thus, as Fig. 20B and 20DAs shown in the figure, even at high concentrations, the measured concentration closely matches the nominal concentration of the prepared sample and is highly linear.

[0516] Example 2: Analytical chromatographic measurements and comparison with UV absorption

[0517] This example demonstrates the use of a QCL-based mid-infrared spectrometer to record absorption spectra essentially in real time to monitor the protein content of the mobile phase eluting from a chromatographic column during elution. Fig.21 The recording of the mid-infrared spectrum over time during elution is shown. Fig.22A and 22B The protein content measured by UV absorbance ( Fig.22A ) is compared with the protein content measured by mid-infrared absorbance ( Fig. 22B , showing the AUC, calculated from the integrated absorbance in the amide-I and amide-II bands, as a function of time). Fig. 22B The mid-infrared data sample quality metric (i.e., AUC) illustrated in the figure shows lower temporal dispersion than the UV data, thus providing more accurate elution monitoring and control.

[0518] Example 3: Amide sub-band deconvolution

[0519] This example demonstrates the use of a QCL-based mid-infrared spectrometer to record the amide band spectra of a protein mixture in aqueous solution and to monitor the change in protein secondary structure with temperature.

[0520] Fig.23A The IR spectra of a 10 mg / mL BSA solution measured while flowing at a rate of 400 μL / min as a function of temperature are shown (each curve was obtained at a different time / temperature). Fig. 23B The second derivative spectra determined from each of the spectra shown in Fig.23A are shown. The second derivative spectra can be used to emphasize / monitor changes in protein secondary structure, which are reflected in the changes in the sub-peaks that make up the amide band region. Fig.23C and 23D show the changes in secondary structure components such as α-helix, turn, β-sheet, and disorder content as a function of temperature. In Fig.23C and 23D , the content of specific secondary structure motifs is measured by determining the absorbance at specific nominal frequencies associated with specific secondary structure motifs. For example, as shown in Fig.23C , the absorbance at 1618 cm -1 is used to measure the β-sheet content, the absorbance at 1692 cm -1 is used to measure the turn content, the absorbance at 1656 cm -1 is used to measure the α-helix content, and the absorbance at 1645 cm -1The absorbance at [location] is used as a measure of the disorder content.

[0521] Another method for monitoring the protein secondary structure content is shown in Fig.24 [reference]. Fig.24 Heat maps are shown that display the variation of the amide sub-band intensities of five different proteins as the temperature increases from room temperature to about 75 °C. To calculate each heat map, rather than monitoring the intensity at a fixed nominal band position, the amide band spectra of each of the five proteins were deconvolved to allow direct determination of the specific peak frequencies and intensities of the constituent sub-bands (e.g., indicative of various secondary structure motifs) from the spectra recorded for each of the five proteins shown in Fig.24 at each temperature.

[0522] Tracking the positions of the sub-bands associated with the secondary structure motifs allows the observation and characterization of the stability of the proteins in different solutions / formulations. For example, comparing the two rightmost heat maps shows the difference in the temperature stability of hen egg white lysozyme (HEWL) protein in phosphate buffer solution (PBS) versus acetate. For example, HEWL in PBS shows a single relatively stable main peak over a wide range of lower temperatures, while the HEWL profile in acetate shows a zigzag band even at lower temperatures, indicating lower stability.

[0523] Thus, this example demonstrates the ability to monitor protein secondary structure changes essentially in real time.

[0524] Example 4: Protein Biophysical Characterization Based on Isoabsorbance Points

[0525] This example demonstrates how to use IR absorption spectra spanning the amide band region and just above the amide band region to calculate a sample mass metric that characterizes the protein density.

[0526] Fig.25 Spectra of various concentrations of BSA are shown. As described herein, the amide band spectra can be used to determine various information about the protein under study. The inset shows an enlarged view of the spectrum near 1700 cm -1 . Without wishing to be bound by any particular theory, relative to the background spectrum of pure water (i.e., having been normalized relative to the background spectrum of pure water), the absorbance at a specific wavenumber of the protein in the water sample can be calculated as shown in Fig.25 , where A is the absorbance at a specific wavenumber, σ p is the absorption cross-section of the protein, γ is the relative density of the protein relative to water, σ w is the absorption cross-section of water, C p (t1) is the concentration of the protein (at the time of measurement), and L is the path length.

[0527] For wavenumbers greater than about 1695 cm -1The wavenumber, the protein absorption cross-section σ p is approximately zero. Thus, above 1695 cm -1 A = -(γσ w )·C p (t1)·L. The protein still affects absorption, not because it absorbs light itself, but because it displaces water. Since σ w is known, the physical constant and C p (t1)·L can be measured (e.g., by integrating over the amide I and II regions), and γ, the relative protein density (relative to water), can be calculated.

[0528] Thus, this example demonstrates how to calculate a measure of protein density (relative to water) from mid-infrared spectra. In certain embodiments, other measures such as hydrodynamic radius and diffusion constant can be calculated.

[0529] Example 5: Baseline correction using a conductivity sensor during elution

[0530] This example shows the baseline change due to the increased presence of ions in the measurements recorded during column elution. Fig.26A The spectral changes before and after elution are shown. Fig.26B The variation of absorption with conductivity is shown.

[0531] Without wishing to be bound by any particular theory, it is believed that Fig.26A the relatively flat shape of the water absorption curve shown in

[0532] Example 6: Protein aggregation metric from IR absorption spectra

[0533] This example demonstrates the calculation of a sample quality metric providing a protein aggregation measurement according to various embodiments described herein.

[0534] The absorption spectra of static or flowing protein aqueous mixtures can be collected in the range from about 800 cm -1 to about 1800 cm -1 or any sub-range thereof, continuously or discontinuously. Such absorption spectra can be obtained using various infrared absorption spectrometers, including, for example, Fourier transform infrared (FT-IR), quantum cascade laser infrared (QCL-IR) analyzers, etc.

[0535] Fig. 27Shows a typical IR absorption spectrum of a protein measured in an aqueous buffer. The absorption spectrum is calculated using a reference background taken from a nominally zero-protein aqueous buffer. Fig. 27 The protein infrared absorption spectrum can be analyzed by one or more of the techniques described in this example, such as Fig. 27 the spectrum shown, to determine sample quality metrics indicative of protein heterogeneity, particularly aggregation (protein aggregation metric) and total protein content. These sample quality metrics can then be used to create digital or analog signals that can be used for dynamic process control and / or process optimization.

[0536] In certain embodiments, sample quality metrics related to protein content, heterogeneity, aggregation, etc. can be calculated based on the characteristics of the absorption peaks associated with the amide-I and / or amide-II bands in the IR spectral data. Table 2 below shows the values of the mid-infrared approximate ranges of the amide-I and / or amide-II spectral bands.

[0537] Fig.28A and 28B shows a first protein aggregation metric calculated as a specific peak metric, namely the "centroid" of the amide II band, which roughly corresponds to the region from about 1500 to about 1600 cm -1 (ν COM酰胺-II ). Fig.28A shows an exemplary spectrum of a zero-aggregation state (e.g., all monomers). Fig.28B shows the spectrum of a 5 mass% aggregation state, which shows that the centroid of the amide II band varies proportionally with the relative concentration of aggregates. The nominal amide II centroid of a high-purity protein (e.g., monoclonal antibody) can be about 1546.2 cm -1 ( Fig.28A ), while protein aggregates or fragments can be above or below this nominal centroid value by plus or minus 0.5 cm -1 ( Fig.28B ). Such small frequency shifts can be easily observed using a high-performance QCL-based infrared mid-infrared analyzer.

[0538] Table 2: Approximate ranges of the amide-I and amide-II bands

[0539]

[0540] Among them, the amide-II centroid frequency has been found to be relatively insensitive to baseline fluctuations during elution, which allows it to be used as a useful metric even without precise baseline correction. In addition, the amide-II centroid frequency has been found to be relatively stable within a small range used to measure monomeric protein species, but deviates from its stable range once aggregated proteins, such as dimers or multimers, are present. Thus, in certain embodiments, the value of the amide-II band can be calibrated for a specific protein (which can be done before or during elution) and then used, for example, as a trigger when it deviates from the stable range to indicate the presence of aggregates.

[0541] Fig.29A and 29B shows a second protein aggregation metric calculated as the ratio of two peak metrics. The first amide-I peak metric characterizes the amide-I band and is calculated as the area under the curve (AUC) of the amide-I band (corresponding to the 1600 to 1700 wavenumber region), and the amide-II peak metric characterizes the amide-II band and is calculated as the area under the curve (AUC) of the amide-II band (roughly corresponding to the 1500 to 1600 wavenumber region). Then the second protein aggregation metric R 酰胺-I比酰胺-II is calculated as the ratio of the amide-I AUC to the amide-II AUC (alternatively, amide-II over amide-I). Typical values of the ratio are provided in Tables 3A and 3B below. High-purity protein monomers typically have a ratio in the range of 1.2 to 1.6. Aggregates typically have a ratio between 0.8 and 1.0, and fragments have a ratio between 0.4 and 0.7.

[0542] Table 3A: Ranges and typical values of four peak metrics

[0543]

[0544] Table 3B: Examples of possible values of four peak metrics

[0545]

[0546]

[0547] The third embodiment ( Fig. 30A and 30B ) involves calculating the peak height ratio of amide I to amide II (alternatively, amide II to amide I), where the amide-I band roughly corresponds to the 1600 - 1700 wavenumber region and the amide-II band roughly corresponds to the 1500 - 1600 wavenumber region. Typical values of the ratio are provided in Fig.26A . High-purity protein monomers typically have a ratio in the range of 1.2 to 1.6. Aggregates typically have a ratio between 0.8 and 1.0, and fragments have a ratio between 0.4 and 0.7.

[0548] Discover the first aggregation metric ν COM酰胺-II Provides the highest correlation with aggregation level and performance, as a quantitative measure of aggregation %.

[0549] Go to Fig.31 , and measure the total protein content using the total area under the curves of the amide-I and amide-II bands.

[0550] Go to Fig.32A and 32B , and then these sample quality metrics can be used to create digital or analog signals that can be used for dynamic process control and / or process optimization. For example, Fig.32B is an illustrative sketch showing the expected (hypothetical) variation over time of sample quality metrics that measure total protein content and protein aggregation as described herein during elution from an IEX column. The solid (black) curve shows the expected variation of the total protein content metric value calculated by integrating over the amide-I and amide-II band regions as Fig.31 described. When monomeric protein begins to elute from the column, the total protein content is expected to peak. The dashed (black) curve shows the protein content measured by UV absorption, which reflects the variation measured by mid-infrared. The dash-dotted (red) curve shows the expected variation of the protein aggregation metric such as ν COM酰胺-II . The protein aggregation metric curve initially peaks at 3202 because low molecular weight species (such as fragments) are eluted first, then reaches and remains at a relatively stable value 3204 for a period of time because high purity monomers are eluted, after which its value shifts to reflect the elution of higher molecular weight species (such as dimers, trimers, etc.) 3208. In some embodiments, as Fig.32B shown, it is expected that using a protein aggregation metric such as ν COM酰胺-II can allow collection during an additional period 3206, during which the total protein content metric measured by IR absorption spectroscopy or UV absorption begins to decrease, but high purity monomers continue to elute, thereby increasing the yield. Additional discussion of process control based on protein aggregation metrics is provided in Example 10 below.

[0551] Example 7: Protein Identification and Metric Development Workflow

[0552] This example shows an illustrative embodiment of an exemplary workflow that can be used to create sample quality metrics and identify reference spectra.

[0553] Fig.33A and 33B show an exemplary method for creating a protein identification method. Fig.33C Provides a schematic diagram showing a non-limiting list of functional modules and sample quality metrics that can be used to evaluate bioproduction process control according to various embodiments described herein.

[0554] Example 8: Downstream Purification Process Monitoring and Control

[0555] This example provides an exemplary control strategy based on the measurement of multiple sample quality metrics for real-time control of collection start / stop based on protein purity. Go to Fig.34A , in an exemplary control strategy, two sample quality metrics are monitored in real time - the total protein content metric and the protein aggregation metric, such as the total amide band AUC and the amide-II centroid, as described in Example 6. These two metrics can be used to calculate an output analog voltage that triggers the start and stop of collecting the desired elution fraction, thereby optimizing the purity of the collection (e.g., in terms of monomers).

[0556] In particular, Fig.34A shows the variation of the voltage signal over time for controlling chromatographic elution in an IEX column. The topmost first trace shows the signal to initiate the conductivity ramp to start elution. The second trace from the top shows the signal to trigger the mid-infrared analyzer to record a reference background spectrum such as water absorbance. As shown, the background spectrum is recorded just before the conductivity ramp starts. After the conductivity ramp starts, a short time period 3402 (light blue shaded area) before the proteins start to elute is shown in the figure while the conductivity is increasing. This time period can be used to calibrate the conductivity-related baseline removal function (e.g., a drainage volume depending on a given conductivity level) if needed. The third trace from the top shows the variation of the analog voltage output for measuring the total protein content. This voltage output can be used, for example, to trigger the start of collection because once the pure monomers start to elute, the voltage output rises and stabilizes. Below the total protein content voltage trace is an analog voltage trace proportional to the amide-II centroid frequency (calculated as described in Example 6 above). As shown, this voltage is stable when mainly monomeric proteins are eluting and then deviates from its set value once the dimers start to elute. Thus, the amide-II centroid voltage can be used to trigger the stop of collection.

[0557] Go to Fig.34B , in certain embodiments, additional sample quality metrics can be calculated and used to further refine and / or optimize the collection window. For example, in certain embodiments, an "other species" protein metric can be calculated. In certain embodiments, sample quality metrics that measure the content of low molecular weight species (e.g., fragments) and high molecular weight species can be calculated and used to generate analog and / or digital control signals.

[0558] Example 9: Downstream Filtration Process Monitoring

[0559] This example demonstrates the use of the systems and methods described herein to perform accurate real-time quantification of proteins from 1 to 300+ g / L, provide real-time quantification and control of buffer stoichiometry, and provide real-time monitoring of protein aggregation and / or structural changes (e.g., denaturation).

[0560] Fig.35 Data generated by several consecutive injections in a TFF system are shown, including an amide-I sub-band deconvolution heat map that can be used to visualize secondary structure changes and total protein concentration. Visualizing secondary structure changes in the filtrate and / or retentate from a TFF system can be used to optimize the filtration or buffer exchange process, e.g., to determine whether protein aggregation or degradation has occurred in the secondary structure.

[0561] Example 10: Wide Spectral Coverage and Low Noise QCL System

[0562] This example demonstrates a QCL mid-infrared analyzer system with improved performance according to certain embodiments described herein. For example, Fig.36 shows the use of multiple QCLs to obtain continuous coverage from about 1025 cm -1 to about 1725 cm -1 . Fig.37 shows the ability to achieve >10-fold sensitivity improvement.

[0563] Example 11: Simultaneous Measurement of Multiple Analytes

[0564] This example demonstrates the simultaneous measurement of multiple analytes in solution by a QCL-based mid-infrared spectrometer according to certain embodiments described herein.

[0565] Fig.38A and 38B show the measured absorbance spectra of several exemplary analytes in solution, and Fig.39A and 39B demonstrate the excellent linearity of glucose in water ( Fig.39A ) and ammonia in water ( Fig.39B ). Fig.40A and 40B show the ability to identify and measure target analytes in complex mixtures (e.g., in the presence of other analytes) with high linearity (RMS concentration error of about 10 micrograms / mL).

[0566] Example 12: Formulation Development and Forced Degradation

[0567] This example demonstrates the use of mid-infrared absorption spectroscopy to observe changes in protein secondary structure over time, e.g., changes related to formulation development.

[0568] Fig.41AShows the variation of the protein amide band spectrum with temperature. Without being bound by any particular theory, in the range of about 1750 cm -1 - 1300 cm -1 The IR spectrum in the range (e.g., when measured with sufficient sensitivity and / or spectral resolution) provides rich information about the intermolecular interactions with the side chains and the NH bending of the amide-II peak.

[0569] Figure 41B Shows Figure 41A A view of the amide-I band region of the spectrum shown, Figure 41C Shows the second derivative spectrum of the amide-I band region, which illustrates the changes in secondary structure, such as the relative decrease in α-helix content and the increase in turn content.

[0570] Figure 42 Shows the spectrum of 10 mg / ml mAb in a specific formulated buffer (blue) and the deconvolution of the constituent sub-bands (red curve).

[0571] Figure 43A Shows the change in the amide-I band structure of mAb with a high β-sheet content as the temperature increases from 25 °C to 80 °C. The central inset is a heat map showing the variation of sub-band intensity with temperature. Figure 43B Summarizes the changes in secondary structure motifs and content as the temperature increases, as shown by the overlapping traces on the heat map.

[0572] Example 13: Data Automation and Stability

[0573] This example demonstrates the automation capabilities, reproducibility, and applicability of the mid-infrared analyzer described in certain embodiments herein, for example, related to GMP-compliant bioproduction.

[0574] Figure 44 Shows an exemplary analysis system with multiple mid-infrared analyzers stacked and linked together and connected to a central computer. In certain embodiments, the mid-infrared system is connected to a processor that allows workflow management of multiple instruments and fluid processors. In certain embodiments, the system provides an automatic self-check for wavelength accuracy and optical power. In certain embodiments, the system includes an alarm system for leaks and / or volatile compounds. In certain embodiments, the system includes an OPC-UA secure communication and instrument control interface, thereby allowing compliance with 21 CFR P11.

[0575] Figures 45A - 45E Shows the results of several reproducibility tests, including high reproducibility of consecutive injections and groups of repeated injections over multiple days.

[0576] Example 14: Independent Protein and Nucleic Acid Quantification and Viral Capsid Titer and Whole Fraction Metrics

[0577] This example demonstrates the independent quantification of protein and nucleic acid content using a QCL-based mid-infrared analyzer and describes several exemplary methods by which protein and nucleic acid content metrics can be used to determine viral capsid concentration and full fractions.

[0578] In particular, as described herein, the amide-I, amide-II, and amide-III bands can be used to measure the protein content in a sample, e.g., based on peak intensity metrics across one or more measurements in the amide-I, amide-II, and amide-III spectral regions, such as peak amplitude or AUC. Similarly, peak intensity metrics of the asymmetric and / or symmetric PO4 bands can be used to determine the nucleic acid content in a sample. In the case of a viral vector sample comprising a protein capsid encapsulating a genetic payload, the protein content metric calculated from the amide-I, II, and / or III spectral bands can thus provide a measure of the total capsid content, while the nucleic acid content can be used to confirm the presence of the genetic payload and compared to the capsid content to determine the full capsid fraction.

[0579] Figure 46A and 46B demonstrates the potential for independent quantification of nucleic acid (DNA in this example) and protein content within a mixture. Figure 46A Absorbance spectra measured from pure samples of nucleic acid and protein, specifically ssDNA and bovine serum albumin (BSA), are plotted. According to various embodiments described herein, the DNA solution spectrum contains absorption peaks within the...

Claims

1. A method for real-time monitoring of protein heterogeneity and (e.g., automated; A method for obtaining a purified sample of a target protein species by controlling a purification process (e.g., semi-automatically), the method comprising: (a) At each of one or more time points, measuring, by one or more mid-infrared (MIR) analyzers, corresponding infrared (IR) absorbance signals of an aqueous sample exiting a purification unit (e.g., a chromatography column), the aqueous sample comprising one or more protein species, the protein species including the target protein species; (b) Receiving, by a processor of a computing device, IR absorbance data corresponding to the IR absorbance signals at each of the one or more time points; (c) Determining, by the processor, values of one or more sample quality metrics based on the IR absorbance data, the one or more sample quality metrics including a protein aggregation metric indicative of the level of protein aggregation within the aqueous sample; and (d) Using the one or more sample quality metrics to control the collection of the target fraction of the aqueous sample (e.g., during a specific collection window) to obtain the purified sample of the target protein species.

2. The method according to claim 1, wherein the purification unit is or comprises a chromatography column.

3. The method according to claim 2, wherein the chromatography column is a member selected from the group consisting of: an affinity chromatography (AC) column, a hydrophobic interaction chromatography (HIC) column, an ion exchange chromatography (IEX) column, a size exclusion chromatography (SEC) column, and a mixed-mode chromatography column [e.g., any combination of the foregoing chromatography columns (e.g., IEX and HIC; e.g., IEX and SEC)].

4. The method according to any one of the preceding claims, wherein the target protein species is or comprises one or more members selected from the group consisting of: monoclonal antibody (mAb), fusion protein, viral capsid protein, antibody-drug conjugate, recombinant protein, and plasma protein.

5. The method according to any one of the preceding claims, wherein the target protein species is or comprises a peptide chain and / or a protein fragment.

6. The method according to any one of the preceding claims, wherein the aqueous sample comprises a plurality of different protein species.

7. The method according to any one of the preceding claims, wherein the aqueous sample comprises a heterogeneous population of the target protein species, which comprises a monomeric portion and an aggregated portion.

8. The method according to any one of the preceding claims, wherein the target protein species is a subspecies of a specific protein species, and the target protein species has a specific desired level and / or type of molecular conjugation (e.g., glycan, small molecule drug, polyethylene glycol, etc.).

9. The method according to any one of the preceding claims, wherein at least one of the one or more MIR analyzers is or comprises a MIR spectrometer, and the MIR spectrometer comprises: A MIR source that is aligned and operable to emit a MIR beam [e.g., including a wavelength range that is substantially within the MIR spectral range (e.g., a range of from about 5000 cm -1 to about 500 cm -1 (e.g., about 2 to 20 microns))]; One or more sampling optics that are aligned to direct and / or allow the MIR beam and / or at least a portion thereof to pass through and / or contact at least a portion of the aqueous sample [e.g., wherein the MIR beam contacts the portion of the aqueous sample by reflection at an interface between a solid material (e.g., an ATR crystal and / or an optical fiber) and the aqueous sample (e.g., wherein the MIR beam undergoes total internal reflection and contacts / detects the portion of the aqueous sample via an evanescent wave extending into the aqueous sample)], and after passing through or contacting the portion of the aqueous sample, towards one or more detectors; and The one or more detectors that are aligned and operable to detect the MIR beam after the MIR beam passes through and / or contacts the aqueous sample.

10. The method according to claim 9, wherein: The one or more sampling optics include a high refractive index material (e.g., an ATR crystal; e.g., an optical fiber) that is aligned such that the MIR beam is directed towards an interface between the high refractive index material and the aqueous sample, incident on the interface, and internally reflected by the interface (e.g., back into the high refractive index material) (e.g., such that the MIR beam is incident on the interface at an angle greater than the critical angle for total internal reflection); and The one or more detectors are aligned and operable to detect the MIR beam exiting the high refractive index material after the MIR beam is internally reflected by the high refractive index material.

11. The method according to claim 10, wherein the high refractive index material is an ATR crystal.

12. The method according to claim 10, wherein the high refractive index material is an optical fiber.

13. The method according to any one of claims 9 to 12, wherein: The one or more sampling optics include a flow cell that includes a detection channel through which the aqueous sample flows; and The one or more detectors are aligned and operable to detect the MIR beam exiting the detection channel after the MIR beam has transmitted through the detection channel.

14. The method according to claim 13, wherein the path length through the detection channel (e.g., followed by the MIR beam during transmission) is about 10 μm or greater (e.g., about or at least 15 μm or greater; e.g., about 25 μm or greater; e.g., about 30 μm or greater; e.g., about 40 μm or greater; e.g., about 50 μm or greater).

15. The method according to any one of claims 9 to 14, wherein the one or more MIR analyzers are or include a quantum cascade laser (QCL)-based MIR spectrometer, which includes a QCL source capable of operating to emit an MIR beam [e.g., including a wavelength range substantially within the MIR spectral range (e.g., a range from about 5000 cm -1 to about 500 cm -1 (e.g., about 2 to 20 microns))].

16. The method according to any one of claims 9 to 15, wherein (e.g., the MIR source is a laser, and) the MIR beam has a spectral linewidth of about 4 cm -1 or less (e.g., about 2 cm -1 or less; e.g., about 1 cm -1 or less; e.g., about 0.5 cm -1 or less).

17. The method according to any one of claims 9 to 16, wherein the power of the MIR beam is about 1 mW or greater (e.g., about 10 mW; e.g., about 50 mW or greater; e.g., about 100 mW or greater; e.g., about 500 mW or greater; e.g., about 1000 mW or greater).

18. The method according to any one of claims 9 to 17, wherein the spectral resolution of the MIR spectrometer is about 4 cm -1 or better (e.g., smaller) [e.g., about 2 cm -1 or better (e.g., smaller); e.g., about 1 cm -1 or better (e.g., smaller); e.g., about 0.5 cm -1 or better (e.g., smaller); e.g., about 0.25 cm -1or better (e.g., smaller); e.g., about 0.1 cm -1 or better (e.g., smaller); e.g., about 0.05 cm -1 or better (e.g., smaller)].

19. The method according to any one of claims 9 to 18, wherein the (e.g., frequency / wavelength) accuracy of the MIR spectrometer is about 2 cm -1 or better (e.g., smaller) [e.g., about 1 cm -1 or better (e.g., smaller); e.g., about 0.5 cm -1 or better (e.g., smaller); e.g., about 0.25 cm -1 or better (e.g., smaller); e.g., about 0.1 cm -1 or better (e.g., smaller); e.g., about 0.01 cm -1 or better (e.g., smaller)].

20. The method according to any one of claims 9 to 19, wherein the (e.g., frequency / wavelength) repeatability of the MIR spectrometer is about 0.5 cm -1 or better (e.g., smaller) [e.g., about 0.25 cm -1 or better (e.g., smaller); e.g., about 0.1 cm -1 or better (e.g., smaller); e.g., about 0.05 cm -1 or better (e.g., smaller); e.g., about 0.001 cm -1 or better (e.g., smaller)].

21. The method according to any one of claims 9 to 20, wherein the MIR source is a tunable laser (e.g., a tunable QCL) (e.g., capable of operating to scan the emission frequency of the MIR beam through a plurality of frequencies within a scan range), and the method comprises, at each of the one or more time points: scanning the emission frequency of the MIR beam within the scan range of the tunable laser, thereby irradiating the aqueous sample at a plurality of emission frequencies; and detecting the MIR beam at each of the plurality of emission frequencies with the one or more detectors (e.g., having (i) been internally reflected by the interface between the high refractive index material and the aqueous sample and / or (ii) transmitted through the detection channel through which the aqueous sample flows), thereby measuring a corresponding infrared (IR) spectrum comprising a plurality of values as the corresponding IR absorption rate signal of the aqueous sample, each of the plurality of values being associated with and representing and / or based on the power detected at a particular one of the plurality of emission frequencies).

22. The method according to claim 21, wherein the plurality of emission wavelengths includes one or more wavelengths within a spectral band in the range of from about 1800 to about 800 cm -1 (e.g., from about 1725 to about 1025 cm -1 ; e.g., from about 1750 to about 1350 cm -1 ; e.g., from about 1725 to about 1375 cm -1 ; e.g., from about 1700 to about 1500 cm -1 ; e.g., from about 1700 to about 1600 cm -1 ; e.g., from about 1700 to about 1000 cm -1 ).

23. The method according to claim 22, wherein the plurality of emission wavelengths includes one or more wavelengths within a spectral band in the range of from about 1600 cm -1 to about 1500 cm -1 .

24. The method according to any one of the preceding claims, wherein the MIR analyzer is an in-line sensor (e.g., as opposed to an off-line or at-line sensor) that measures the IR absorbance signal substantially in real-time as the aqueous solution exits the purification unit.

25. The method according to any one of the preceding claims, wherein for each of the one or more time points, the IR absorbance data includes a corresponding amide band spectrum [e.g., for each specific wavelength among a plurality of sampled (e.g., emission) wavelengths in the range of from about 1800 to about 800 cm -1 , the amide band spectrum includes a relevant absorbance value representing the absorption level of a portion of the aqueous sample at the specific wavelength].[[]END]] 26. The method according to claim 25, wherein step (d) includes determining a corresponding value of the protein aggregation metric for each specific time point among at least a portion of the one or more time points.

27. The method according to claim 26, wherein for each specific time point, determining the corresponding value of the protein aggregation metric includes: Calculating a value of an amide II peak metric from the amide band spectrum corresponding to the specific time point, the amide II peak metric quantifying one or more characteristics (e.g., frequency position, line width, intensity) of the amide II band at the specific time point; and Using the value of the amide II peak metric to determine a corresponding value of the protein aggregation metric (e.g., where the protein aggregation metric is the amide II peak metric or a function of the amide II peak metric).

28. The method according to claim 27, wherein the amide II peak metric is a frequency position metric that quantifies the frequency at which the amide II band is substantially centered at the specific time point [e.g., the centroid frequency, the frequency of the maximum height of the amide II band, the center frequency of a fitted peak function (e.g., Gaussian, Lorentz, etc.), etc.].[[]END]] 29. The method according to claim 28, wherein the frequency position metric is the centroid frequency of the amide II band.

30. The method according to claim 26, wherein for each particular time point, determining the corresponding value of the protein aggregation metric comprises: Calculating a value of an amide I peak metric from the amide band spectrum corresponding to the specific time point, the amide I peak metric quantifying one or more characteristics (e.g., frequency position, line width, intensity) of the amide I band at the specific time point; And Using both the value of the amide I peak metric and the value of the amide II peak metric to determine a corresponding value of the protein aggregation metric (e.g., where the protein aggregation metric is a function of the amide I peak metric and the amide II peak metric).

31. The method according to claim 30, wherein: The amide I peak metric is a peak intensity metric that quantifies the intensity of the amide I band at the specific time point (e.g., the peak height of the amide I band, the area under the curve (AUC) of the amide I band); The amide II peak metric is a peak intensity metric that quantifies the intensity of the amide II band at the specific time point (e.g., the peak height of the amide I band, the area under the curve (AUC) of the amide I band); And Determining the value of the protein aggregation metric includes calculating (i) the ratio of the value of the amide I peak metric to the value of the amide II peak metric and / or (ii) the ratio of the value of the amide II peak metric to the value of the amide I peak metric.

32. The method according to any one of the preceding claims, wherein the one or more sample quality metrics further comprise a total protein content metric indicative of the level of protein content within the aqueous sample.

33. The method according to any one of the preceding claims, wherein step (d) comprises causing, by the processor, one or more trigger signals (e.g., voltages) to be transmitted to the controller unit of the purification unit.

34. The method according to claim 33, wherein the one or more trigger signals comprise an analog voltage signal having a time-varying amplitude based on the value of the protein aggregation metric (e.g., substantially proportional thereto).

35. The method according to claim 33 or 34, wherein the one or more trigger signals comprise an analog voltage signal having a time-varying amplitude based on the value of the total protein content metric (e.g., substantially proportional thereto).

36. The method according to any one of claims 33 to 35, wherein step (d) comprises one or both of the following: initiating, by the controller unit, collection of the target fraction of the aqueous sample based on the one or more trigger signals [e.g., wherein a particular one of the one or more trigger signals is an analog signal, and the controller unit initiates collection of the target fraction based on the amplitude of the analog signal (e.g., when it exceeds a particular threshold; e.g., when it drops below a particular threshold); e.g., wherein a particular one of the one or more trigger signals is a digital signal that triggers (e.g., by transitioning from a 0 voltage level to a 1 voltage level, or vice versa) the initiation of collection of the target fraction], and The controller unit stops the collection of the target fraction of the aqueous sample based on the one or more trigger signals [e.g., where a particular one of the one or more trigger signals is an analog signal, and the controller unit stops the collection of the target fraction based on the amplitude of the analog signal (e.g., when it exceeds a particular threshold; e.g., when it drops below a particular threshold); e.g., where a particular one of the one or more trigger signals is a digital signal that triggers (e.g., by transitioning from a 0 voltage level to a 1 voltage level, or vice versa) the stop of the collection of the target fraction].

37. A method for real-time monitoring of protein aggregation in a sample, the method comprising: (a) A processor of a computing device repeatedly receives IR absorption rate data corresponding to IR absorption rate signals measured at each of a plurality of time points. For each specific time point of the plurality of time points, the IR absorption rate data includes a corresponding IR absorption rate spectrum measured from the sample at the specific time point and includes a plurality of absorption rate values, each absorption rate value associated with a specific wave number; (b) For each specific time point of the plurality of time points, the processor (e.g., automatically) analyzes the IR absorption rate data to obtain a real-time protein aggregation signal that provides a measure of protein aggregation in the sample over time by: Using the IR absorption rate spectrum corresponding to the specific time point to determine values of one or more peak metrics of one or both of the amide I band and the amide II band; Using the values of the one or more peak metrics to determine a value of a protein aggregation metric indicative of the level of protein aggregation in the sample at the specific time point; And Updating the real-time protein aggregation signal with the determined value of the protein aggregation metric at the specific time point; And (c) The processor stores and / or provides the real-time protein aggregation signal for one or more of the following purposes: (i) further processing, (ii) display, and (iii) use as a control signal for adjusting one or more purification units (e.g., a chromatography system).

38. The method according to claim 37, wherein step (b) comprises, for each particular time point, determining the value of a frequency position metric as the value of the protein aggregation metric indicating the level of protein aggregation within the sample at the particular time point, the frequency position metric quantifying the frequency at which the amide II band is substantially centered at the particular time point [e.g., centroid frequency, the frequency of the maximum height of the amide II band, the center frequency of a fitted peak function (e.g., Gaussian, Lorentzian, etc.), etc.].

39. The method according to claim 38, wherein the frequency position metric is the centroid frequency of the amide II band.

40. The method according to any one of claims 37 to 39, wherein for each particular time point, step (b) comprises: Determine a value of an amide I peak intensity metric that quantifies the intensity of the amide I band at the particular time point (e.g., the peak height of the amide I band, the area under the curve (AUC) of the amide I band); Determine a value of an amide II peak intensity metric that quantifies the intensity of the amide II band at the particular time point (e.g., the peak height of the amide I band, the area under the curve (AUC) of the amide I band); and Determine (i) the ratio of the amide I peak metric value to the amide II peak metric value and / or (ii) the ratio of the amide II peak metric value to the amide I peak metric value as the value of the protein aggregation metric.

41. A method for monitoring and controlling a production unit for manufacturing a biological product (e.g., a protein; e.g., a virus) based on mid-infrared (MIR) spectroscopy, the method comprising: (a) Measure the corresponding infrared (IR) absorption rate signals of the aqueous samples flowing into and / or out of the production unit (e.g., and which include one or more inputs, output products, waste products, or in-process products of the production unit) at each of one or more time points by one or more (e.g., integrated) mid-infrared (MIR) analyzers [e.g., an MIR analyzer as described in any one of claims 9 to 24]; (b) Receive, by a processor of a computing device, the IR absorption rate data corresponding to the IR absorption rate signals at each of the one or more time points; and (c) Use the received IR absorption rate data to adjust one or more process parameters of: (i) the production unit and / or (ii) a second (e.g., upstream and / or downstream) production unit.

42. The method according to claim 41, wherein the production unit is a purification unit.

43. The method according to claim 42, wherein the purification unit is a member selected from the group consisting of: an alternating tangential flow filtration (ATF) system, a tangential flow depth filtration (TFDF) system, a tangential flow filtration (TFF) system, a chromatography column, a direct or conventional flow filtration unit, an ultrafiltration unit, and a diafiltration unit.

44. The method according to claim 43, wherein the purification unit is a chromatography column {e.g., and wherein the chromatography column is a member selected from the group consisting of: an affinity chromatography (AC) column, a hydrophobic interaction chromatography (HIC) column, an ion exchange chromatography (IEX) column, a size exclusion chromatography (SEC) column, and a mixed mode chromatography column [e.g., any combination of the foregoing chromatography columns (e.g., IEX and HIC; e.g., IEX and SEC)]}.

45. The method according to claim 41, wherein the production unit is a bioreactor (e.g., a seed bioreactor; e.g., a production bioreactor).

46. The method according to any one of claims 41 to 45, wherein the aqueous sample comprises one or more protein species selected from the group consisting of: monoclonal antibodies (mAbs), fusion proteins, viral capsid proteins, antibody-drug conjugates, recombinant proteins, and plasma proteins.

47. The method according to any one of claims 41 to 46, wherein the aqueous sample comprises peptide chains and / or protein fragments.

48. The method according to any one of claims 41 to 47, wherein the aqueous sample comprises a plurality of different protein species.

49. The method according to any one of claims 41 to 48, wherein the aqueous sample comprises a heterogeneous population of a target protein species, which comprises monomeric and aggregated moieties.

50. The method according to any one of claims 41 to 49, wherein the aqueous sample comprises one or more subspecies of a specific protein species, which have a specific desired level and / or type of molecular conjugation (e.g., glycan, small molecule drug, polyethylene glycol, etc.).

51. The method according to any one of claims 41 to 50, wherein the aqueous sample comprises nucleic acids (e.g., DNA, RNA, mRNA, etc.).

52. The method according to any one of claims 41 to 51, wherein the aqueous sample comprises one or more viruses and / or virus-like particles [e.g., adeno-associated virus vectors (AAV); e.g., lentiviral vectors].

53. The method according to any one of claims 41 to 52, wherein step (c) comprises using the IR absorption rate data to determine the value of one or more sample quality metrics at each of the one or more time points (e.g., and adjusting the one or more process parameters based on the value of the one or more sample quality metrics).

54. The method according to claim 53, wherein the one or more sample quality metrics include a total protein content metric that quantifies the amount and / or concentration of protein in the aqueous sample.

55. The method according to claim 53 or 54, wherein the one or more sample quality metrics include a protein aggregation metric that indicates the level of protein aggregation in the aqueous sample.

56. The method according to any one of claims 53 to 55, wherein the one or more sample quality metrics include one or more protein species metrics that identify the presence of one or more specific protein species in the aqueous sample and / or quantify the content of one or more specific protein species in the aqueous sample (e.g., absolute content; e.g., relative content).

57. The method according to any one of claims 53 to 56, wherein the one or more sample quality metrics include a protein conjugation metric that quantifies the level and / or type of molecular conjugation (e.g., glycan, small molecule drug, polyethylene glycol, etc.).

58. The method according to any one of claims 53 to 57, wherein the one or more sample quality metrics include one or more protein secondary structure metrics that quantify the presence and / or content of one or more protein secondary structure motifs (e.g., α-helix content, β-sheet content, turn content, disordered content).

59. The method according to any one of claims 53 to 58, wherein the one or more sample quality metrics include one or more nucleic acid content metrics that quantify the nucleic acid content in the aqueous sample [e.g., total content of nucleic acid (e.g., concentration (e.g., titer), mass per volume (e.g., mg / mL), number of particles per volume, viral genome copy number, etc.); e.g., total content and / or relative content of one or more specific types of nucleic acid (e.g., DNA, RNA, ssDNA, dsDNA)].

60. The method according to any one of claims 53 to 59, wherein the one or more sample quality metrics include one or more virus content metrics that quantify the content of virus and / or virus-like particles in the aqueous sample [e.g., concentration (e.g., titer), mass per volume (e.g., mg / mL), number of (virus) particles per volume, etc.].

61. The method according to any one of claims 53 to 60, wherein the one or more sample quality metrics include one or more empty / full metrics that quantify the content and / or relative fraction of empty viral vectors and / or full viral vectors within the aqueous sample (e.g., the percentage, ratio, etc. of full viral vectors).

62. The method according to any one of claims 53 to 61, wherein the one or more sample quality metrics include a capsid aggregation metric that indicates the level of capsid aggregation within the viral vector sample.

63. The method according to any one of claims 53 to 62, wherein the one or more sample quality metrics include a viral nucleic acid (e.g., viral DNA, RNA, etc.) content metric that differentiates viral nucleic acid from host cell protein and host cell nucleic acid content.

64. The method according to any one of claims 53 to 62, wherein at least a portion (e.g., one or more) of the sample quality metrics are calculated based on one or more peak metrics that measure characteristics of one or more absorption bands in the IR spectral data {e.g., wherein each peak metric is associated with one or more specific spectral bands [e.g., a continuous range of wavelengths / wavenumbers (e.g., amide-I band, amide-II band, amide-III band; e.g., an amide region spanning two or more of the amide bands; e.g., an asymmetric PO4 band; e.g., a symmetric PO4 band)], and quantify one or more specific structural characteristics of one or more absorption peaks within the specific spectral band [e.g., intensity (e.g., peak amplitude; e.g., area under the curve (AUC)); e.g., linewidth; e.g., frequency position (e.g., peak frequency; e.g., centroid frequency)]}.

65. The method according to any one of claims 53 to 64, wherein the IR absorbance data includes: (i) One or more (e.g., multiple) amide II absorption rate values, which are associated with and measure the IR absorption of (a viral sample) at a wavenumber within the amide II spectral band (e.g., in the range of about 1500 cm -1 to about 1600 cm -1 , e.g., in the range of about 1500 cm -1 to about 1575 cm -1 ; e.g., in the range of about 1500 cm -1 to about 1550 cm -1 ; e.g., in the range of about 1540 cm -1 to about 1560 cm -1 ); and / or (ii) one or more (e.g., a plurality of) amide III absorbance values, which are associated with and measure the IR absorbance of (said viral sample) at a wavenumber within the amide II spectral band (e.g., in the range of about 1250 cm -1 to about 1350 cm -1 , e.g., in the range of about 1250 cm -1 to about 1325 cm -1 ; e.g., in the range of about 1275 cm -1 to about 1325 cm -1 ; e.g., in the range of about 1280 cm -1 to about 1300 cm -1 ).

66. The method according to claim 65, wherein determining the one or more sample quality metrics includes determining the value of a protein content metric [e.g., concentration (e.g., titer)] at least in part based on the amide II and / or amide III absorbance values, the protein content metric quantifying the protein content within the sample.

67. The method according to claim 66, comprising: Determine a value of an amide II peak metric based on the amide II absorption rate value and / or determine a value of an amide III peak metric based on the amide III absorption rate value; and Use the amide II peak metric value and / or the amide III peak metric value to determine the protein content metric value.

68. The method according to claim 67, wherein the amide II peak measurement and / or the amide III peak measurement are peak intensity measurements that respectively quantify the intensity of the amide II band and / or the amide III band [e.g., peak height, area under the curve (AUC), etc.].

69. The method according to any one of claims 53 to 68, wherein the IR absorption rate data includes: (i) One or more (e.g., a plurality of) antisymmetric phosphate stretching (antisymmetric - PO4) absorbance values, which are associated with and measure the IR absorption of (the viral sample) at a wavenumber within the antisymmetric - PO4 spectral band (e.g., in the range of about 1150 cm -1 to about 1250 cm -1 ; e.g., in the range of about 1175 cm -1 to about 1250 cm -1 ; e.g., in the range of about 1200 cm -1 to about 1250 cm -1 ; e.g., in the range of about 1210 cm -1 to about 1230 cm -1 ); and / or (ii) one or more (e.g., a plurality of) symmetric phosphate stretching (symmetric-PO4) absorbance values, which are associated with and measure the IR absorbance of (the viral sample) at a wavenumber within the symmetric-PO4 spectral band (e.g., in the range of about 1000 cm -1 to about 1100 cm -1 ; e.g., in the range of about 1050 cm -1 to about 1100 cm -1 ; e.g., in the range of about 1075 cm -1 to about 1100 cm -1 ; e.g., in the range of about 1075 cm -1 to about 1085 cm -1 ).

70. The method according to claim 69, wherein determining the value of the one or more sample quality metrics includes determining a value of a nucleic acid (e.g., DNA, RNA, etc.) content metric that at least partially is based on the antisymmetric-PO4 and / or symmetric-PO4 absorption rate values, the nucleic acid content metric quantifying the nucleic acid content within the sample [e.g., concentration (e.g., titer)].

71. The method according to claim 70, which includes: Determine a value of an antisymmetric-PO4 peak metric based on the antisymmetric-PO4 absorption rate value and / or determine a value of a symmetric-PO4 peak metric based on the symmetric-PO4 absorption rate value; and Use the antisymmetric-PO4 peak metric value and / or the symmetric-PO4 peak metric value to determine the nucleic acid content metric value.

72. The method according to claim 71, wherein the antisymmetric-PO4 peak measurement and / or the symmetric-PO4 peak measurement are peak intensity measurements that respectively quantify the intensity of the antisymmetric-PO4 band and / or the symmetric-PO4 band [e.g., peak height, area under the curve (AUC), etc.].

73. The method according to any one of claims 53 to 72, wherein determining the value of the one or more sample quality metrics includes determining (i) a value of a protein content metric that quantifies the protein content within the sample [e.g., concentration (e.g., titer)] and (ii) a value of a nucleic acid (e.g., DNA, RNA, etc.) content metric that quantifies the nucleic acid content within the sample [e.g., concentration (e.g., titer)], thereby independently quantifying the total protein and nucleic acid content within the sample.

74. The method according to claim 73, wherein determining the value of the one or more sample quality metrics includes determining the total capsid content at least partially based on the value of the protein content metric (e.g., in the form of a function of the value of the protein content metric).

75. The method according to claim 73 or 74, wherein determining the value of the one or more sample quality metrics includes determining the full capsid fraction at least partially based on (i) the value of the protein content metric and / or the total capsid content and (ii) the value of the nucleic acid content metric (e.g., in the form of a function of (i) and (ii)).

76. The method according to any one of claims 53 to 75, wherein the IR absorption rate data is or includes one or more IR absorption rate spectra, and for each specific wavenumber among a plurality of wavenumbers spanning the measured spectral band, each IR absorption rate spectrum includes a corresponding IR absorption rate value, and the corresponding IR absorption rate value represents a measure of the absorption of IR light by the aqueous sample at the specific wavenumber.

77. The method according to claim 76, wherein the measured spectral band spans one or more bands selected from the group consisting of: amide II band, amide III band, asymmetric - PO4 band, and symmetric - PO4 band.

78. The method according to any one of claims 53 to 77, which includes determining the value of the one or more sample mass metrics at each of the one or more time points, thereby monitoring the one or more sample mass metrics over time.

79. The method according to claim 78, which includes determining the value of the one or more sample mass metrics substantially in real - time.

80. The method according to any one of claims 53 to 79, wherein a machine - learning model is used to calculate at least one specific sample mass metric among the one or more sample mass metrics, and the machine - learning model receives one or more IR spectra as inputs and generates the specific sample mass metric as an output.

81. The method according to any one of claims 53 to 80, wherein calculating the specific sample mass metric includes de - convolving the amide spectral region into sub - bands and / or calculating a second - derivative spectrum.

82. The method according to any one of claims 53 to 81, wherein the IR absorption rate data includes IR absorption rate spectra, and wherein step (c) includes: Receive (e.g., and / or access) one or more reference spectra, each reference spectrum measured from a corresponding (e.g., high-quality) reference sample [e.g., including a target viral vector species of high purity and / or concentration and / or one or more model components thereof (e.g., a model protein solution; e.g., a model ssDNA solution)]; and Use the IR absorption rate spectrum and the one or more reference spectra (e.g., automatically) to determine at least a portion of the values of the one or more sample quality metrics [e.g., determine one or more (e.g., multiple) measures of the deviation between the reference spectrum and the IR absorption rate spectrum as the portion of the values of the one or more viral vector sample quality metrics].

83. The method according to claim 82, wherein the one or more reference spectra include high - quality virus - vector spectra measured from a reference sample that has a full - capsid fraction equal to or higher than a specific threshold fraction (e.g., known a priori; e.g., determined to be).

84. The method according to claim 83, wherein the threshold fraction is about 75% [e.g., about 80% (e.g., about 90%)].

85. The method according to any one of claims 82 to 84, wherein step (c) comprises calculating a difference spectrum based on at least one of the one or more reference spectra and the IR absorption rate spectrum (e.g., by subtracting the IR absorption rate spectrum and / or a scaled or otherwise pre - processed version thereof from the reference spectrum and / or a scaled or otherwise pre - processed version thereof, or vice versa).

86. The method according to any one of claims 82 to 85, wherein step (c) comprises calculating one or more derivative spectra of at least one of the one or more reference spectra and / or the IR absorption rate spectrum (e.g., first - order derivative; e.g., second - order derivative).

87. The method according to any one of claims 82 to 86, wherein step (c) comprises calculating (e.g., as one or more of the sample quality metrics values) one or more members selected from the group consisting of: a correlation value based on the correlation between (i) a particular one of the one or more reference spectra and / or one or more of its derivatives and (ii) the IR absorption rate spectrum and / or one or more of its derivatives; a covariance value based on the covariance between (i) a particular one of the one or more reference spectra and / or one or more of its derivatives and (ii) the IR absorption rate spectrum and / or one or more of its derivatives; (i) The Pearson's correlation value between a specific one of the one or more reference spectra and / or one or more of its derivatives and (ii) the IR absorption rate spectrum and / or one or more of its derivatives; and The overlapping integral value based on the overlapping integral of (i) a specific one of the one or more reference spectra and / or one or more of its derivatives and (ii) the IR absorption rate spectrum and / or one or more of its derivatives.

88. The method according to any one of claims 82 to 87, wherein step (c) comprises: Determine the values of a set of one or more specific peak metrics from the IR absorption rate spectrum, thereby obtaining a set of sample peak metric values; and Determine the values of one or more of the sample quality metrics based on the set of sample peak metric values and a set of reference peak metric values that have been determined for the one or more specific peak metrics from the one or more reference spectra.

89. The method according to any one of claims 82 to 88, which comprises determining (e.g., as one or more of the sample quality metric values) a similarity score that measures the similarity between the one or more reference spectra and the IR absorption rate spectrum.

90. The method according to any one of claims 82 to 89, which comprises repeating steps (a) - (d) substantially in real - time, thereby monitoring in real - time the deviation from the one or more reference spectra.

91. The method according to any one of claims 53 to 90, wherein the one or more time points are a plurality of time points [e.g., step (a) comprises (e.g., repeatedly) measuring the IR absorption signal at each of the plurality of time points (e.g., continuously, in real - time)].

92. The method according to claim 91, comprising determining a value of a first sample quality metric at each of the plurality of time points, and determining a value of a second (e.g., time difference; e.g., time aggregation) sample quality metric using the values of the first sample quality metric corresponding to two or more of the plurality of time points.

93. The method according to claim 92, wherein the second sample quality metric is a time difference metric that measures a temporal change of the first sample quality metric, and is calculated based on a difference between (i) the value of the first sample quality metric of a first set of time points and (ii) the value of the first sample quality metric of a second set of time points [e.g., a difference between the value of the first sample quality metric at a first (e.g., current) time point and the value of the first sample quality metric at a second (e.g., previous) time point (e.g., a difference between values of consecutive time points)].

94. The method according to claim 92 or 93, wherein the second sample quality metric is a (e.g., real-time) time aggregation signal that is a function (e.g., a running sum, average, median, mode, variance, standard deviation, etc. over a specific time window) of at least a portion of the plurality of time points [e.g., a cumulative increasing portion, e.g., starting at a specific time point and ending at the current time point; e.g., a time window of a specific size (e.g., a look-back window)].

95. The method according to any one of claims 41 to 94, wherein step (c) includes causing, by the processor, one or more trigger signals (e.g., voltages) to be transmitted to a controller unit of the production unit.

96. The method according to claim 95, wherein the one or more trigger signals include an analog voltage signal having a time-varying amplitude based on at least a portion of the value of one or more sample quality metrics (e.g., being substantially proportional thereto).

97. The method according to any one of claims 41 to 96, wherein step (c) includes using a machine learning model to adjust the one or more process parameters [e.g., wherein the machine learning model receives one or more sample quality metrics as inputs and generates an adjustment to a target process parameter and / or a target process parameter as an output; e.g., wherein the machine learning model receives one or more IR spectra as inputs and generates an adjustment to a target process parameter and / or a target process parameter as an output].

98. The method according to any one of claims 41 to 97, wherein the one or more process parameters include one or more members selected from the group consisting of flow rate, flow direction, pressure, temperature, and pH.

99. The method according to any one of claims 41 to 98, wherein the one or more process parameters include the amount (e.g., absolute amount and / or relative amount) of one or more raw materials (e.g., used as the input to the production unit).

100. The method according to any one of claims 41 to 99, wherein the one or more process parameters include the time to initiate and / or stop a subprocess (e.g., heating, collecting elution fractions, growth, etc.).

101. The method according to any one of claims 41 to 100, wherein: When the aqueous sample exits the purification unit (e.g., chromatography column), measure one or more IR absorption rate signals corresponding to the IR absorption rate data received in step (b) from it (the aqueous sample) at each of one or more time points; and The method includes: Using the IR absorption data to determine one or both of (i) the total capsid content and (ii) the full capsid fraction; and Using the determined total capsid content and / or full capsid fraction to control the collection of the target fraction of the aqueous sample (e.g., during a specific collection window), thereby obtaining a purified sample of the viral vector material.

102. The method according to any one of claims 41 to 101, wherein: When the aqueous sample exits the purification unit (e.g., chromatography column), measure one or more IR absorption rate signals corresponding to the IR absorption rate data received in step (b) from it (the aqueous sample) at each of one or more time points; and The method includes: Using the IR absorption data to determine a capsid aggregation metric that measures the level of aggregation between capsids in the viral vector sample; and Using the capsid aggregation metric to control the collection of the target fraction of the aqueous sample (e.g., during a specific collection window), thereby obtaining a purified sample of the viral vector material.

103. A system for real-time monitoring of protein heterogeneity and (e.g., automated; For example, a (semi - automated) system for controlling a purification process to obtain a purified sample of a target protein species, the system includes: (a) One or more mid - infrared (MIR) analyzers that are aligned and operable to measure, at each of one or more time points, the corresponding infrared (IR) absorption rate signal of an aqueous sample exiting a purification unit (e.g., chromatography column), the aqueous sample including one or more protein species that include the target protein species; (b) A processor of a computing device; and (c) A memory having instructions stored thereon, wherein the instructions, when executed by the processor, cause the processor to: Receive IR absorption rate data corresponding to the IR absorption rate signal at each of the one or more time points; Determine the values of one or more sample quality metrics based on the IR absorption rate data, the one or more sample quality metrics including a protein aggregation metric indicating the level of protein aggregation within the aqueous sample; Provide (e.g., transmit to a controller unit of the purification unit) and / or use the one or more sample quality metrics to control the collection of the target fraction of the aqueous sample (e.g., during a specific collection window) to obtain a purified sample of the target protein species.

104. The system according to claim 103, further comprising the purification unit and / or its controller unit.

105. The system according to claim 104, wherein the purification unit is or comprises a chromatography column.

106. A system for real-time monitoring of protein aggregation in a sample, the system comprising: A processor of a computing device; And A memory having instructions stored thereon, wherein the instructions, when executed by the processor, cause the processor to: (a) Repeatedly receive IR absorbance data corresponding to IR absorbance signals measured at each of a plurality of time points, and for each specific time point of the plurality of time points, the IR absorbance data includes a corresponding IR absorbance spectrum measured from the sample at the specific time point and includes a plurality of absorbance values, each absorbance value associated with a specific wave number; (b) For each specific time point of the plurality of time points, analyze the IR absorbance data (e.g., automatically) to obtain a real-time protein aggregation signal that provides a measure of the amount of protein aggregation in the sample over time: Use the IR absorbance spectrum corresponding to the specific time point to determine the value of one or more peak metrics in the amide I band and / or the amide II band; Use the value of the one or more peak metrics to determine the value of a protein aggregation metric indicative of the level of protein aggregation in the sample at the specific time point; And Update the real-time protein aggregation signal with the determined value of the protein aggregation metric at the specific time point; And (c) Store and / or provide the real-time protein aggregation signal for one or more of the following uses: (i) Further processing, (ii) Display, and (iii) Use as a control signal for adjusting one or more purification units (e.g., a chromatography system).

107. A system for monitoring and controlling a production unit for manufacturing biological products (e.g., proteins; e.g., viruses) based on mid-infrared (MIR) spectroscopy, the method comprising: (a) One or more (e.g., integrated) mid-infrared (MIR) analyzers [e.g., the MIR analyzer as claimed in any one of claims 9 to 24], which are aligned and operable to measure corresponding infrared (IR) absorbance signals of an aqueous sample flowing into and / or out of the production unit (e.g., and which includes one or more inputs, output products, waste products, or in-process products of the production unit) at each of one or more time points; (b) A processor of a computing device; And (c) A memory having instructions stored thereon, wherein the instructions, when executed by the processor, cause the processor to: Receive IR absorbance data corresponding to the IR absorbance signals at each of the one or more time points; and Use the received IR absorbance data to cause an adjustment of one or more process parameters of the production unit.

108. The system according to claim 107, further comprising the production unit and / or its controller unit.

109. A method for quantifying and / or monitoring the quality of viral vectors in an aqueous sample comprising one or more viruses and / or virus-like particles (e.g., in real time), the method comprising: (a) Receive (e.g., repeatedly) by a processor of a computing device IR absorbance data corresponding to one or more infrared (IR) absorbance signals measured from the sample; (b) Determine by the processor the value of one or more virus vector sample quality metrics using the IR absorbance data (e.g., automatically); And (c) Storing and / or providing the one or more viral vector sample quality metrics for display and / or further processing.

110. The method according to claim 109, wherein the one or more viral vector quality metrics include quantifying the total capsid content of the viral capsid content in the sample [e.g., concentration (e.g., titer), mass per volume (e.g., mg / mL), number of (viral) particles per volume, etc.].

111. The method according to claim 109 or claim 110, wherein the one or more viral vector sample quality metrics include the full capsid fraction (e.g., percentage, ratio, etc. of full viral vectors).

112. The method according to any one of claims 109 to 111, wherein the one or more viral vector sample quality metrics include a capsid aggregation metric that indicates the level of capsid aggregation in the viral vector sample.

113. The method according to any one of claims 109 to 112, wherein the one or more viral vector sample quality metrics include a protein content metric that quantifies the protein content in the sample [e.g., concentration (e.g., titer), mass per volume (e.g., mg / mL), number of particles per volume, etc.].

114. The method according to any one of claims 109 to 113, wherein the one or more viral vector sample quality metrics include nucleic acid (e.g., DNA, RNA, etc.) content metrics that quantify the nucleic acid content within the sample [e.g., concentration (e.g., titer), mass per volume (e.g., mg / mL), number of particles per volume, viral genome copy number, etc.].

115. The method according to any one of claims 109 to 114, wherein the one or more viral vector sample quality metrics include viral nucleic acid (e.g., viral DNA, RNA, etc.) content metrics that distinguish viral nucleic acid from host cell protein and host cell nucleic acid content.

116. The method according to any one of claims 109 to 115, wherein step (b) comprises: Determining, by the processor, the value of each of the one or more peak metrics of the IR absorption data, wherein each peak metric is associated with one or more specific spectral bands [e.g., a continuous range of wavelengths / wavenumbers (e.g., amide-I band, amide-II band, amide-III band; e.g., an amide region spanning two or more of the amide bands; e.g., asymmetric PO4 band; e.g., symmetric PO4 band)] and quantifying a specific structural feature of one or more absorption peaks within the specific spectral band [e.g., intensity (e.g., peak amplitude; e.g., area under the curve (AUC)); e.g., linewidth; e.g., frequency position (e.g., peak frequency; e.g., centroid frequency)]; and Using the determined values of the one or more peak metrics to determine at least a portion of the value of the viral vector sample quality metric.

117. The method according to any one of claims 109 to 116, wherein the IR absorption rate data includes: (i) One or more (e.g., a plurality of) amide II absorbance values, which are associated with and measure the IR absorption of (the viral sample) at a wavenumber within the amide II spectral band (e.g., in the range of about 1500 cm -1 to about 1600 cm -1 , e.g., in the range of about 1500 cm -1 to about 1575 cm -1 ; e.g., in the range of about 1500 cm -1 to about 1550 cm -1 ; e.g., in the range of about 1540 cm -1 to about 1560 cm -1 ); and / or (ii) One or more (e.g., multiple) amide III absorbance values, which are associated with and measure the IR absorbance of (the viral sample) at a wavenumber within the amide II spectral band (e.g., in the range of about 1250 cm -1 to about 1350 cm -1 , e.g., in the range of about 1250 cm -1 to about 1325 cm -1 ; e.g., in the range of about 1275 cm -1 to about 1325 cm -1 ; e.g., in the range of about 1280 cm -1 to about 1300 cm -1 ).

118. The method according to claim 117, wherein step (b) comprises determining a value of a protein content metric [e.g., concentration (e.g., titer)] at least in part based on the amide II and / or amide III absorption rate values, the protein content metric quantifying the protein content within the sample.

119. The method according to claim 118, which comprises: Determining the value of the amide II peak metric based on the amide II absorbance value and / or determining the value of the amide III peak metric based on the amide III absorbance value; And Using the amide II peak metric value and / or the amide III peak metric value to determine the protein content metric.

120. The method according to claim 119, wherein the amide II peak metric and / or the amide III peak metric are peak intensity metrics that respectively quantify the intensity [e.g., peak height, area under the curve (AUC), etc.] of the amide II band and / or the amide III band.

121. The method according to any one of claims 109 to 120, wherein the one or more viral vector sample quality metrics include one or more protein structure metrics (e.g., protein structure metrics; e.g., protein tertiary and / or quaternary structure metrics) that indicate the presence and / or content (e.g., absolute content; e.g., relative content) of one or more specific protein structure forms (e.g., specific secondary structure motifs; e.g., specific tertiary and / or quaternary structure motifs / forms) within the sample (e.g., thereby providing monitoring of changes in the secondary / tertiary / quaternary structure of the capsid protein).

122. The method according to any one of claims 109 to 121, wherein the IR absorption rate data includes: (i) One or more (e.g., multiple) antisymmetric phosphate stretching (antisymmetric - PO4) absorbance values, which are associated with and measure the IR absorbance of (the viral sample) at wavenumbers within the antisymmetric - PO4 spectral band (e.g., in the range of about 1150 cm -1 to about 1250 cm -1 ; e.g., in the range of about 1175 cm -1 to about 1250 cm -1 ; e.g., in the range of about 1200 cm -1 to about 1250 cm -1 ; e.g., in the range of about 1210 cm -1 to about 1230 cm -1 ); and / or (ii) one or more (e.g., a plurality of) symmetric phosphate stretching (symmetric-PO4) absorbance values, which are associated with and measure the IR absorbance of (the viral sample) at a wavenumber within the symmetric-PO4 spectral band (e.g., in the range of about 1000 cm -1 to about 1100 cm -1 ; e.g., in the range of about 1050 cm -1 to about 1100 cm -1 ; e.g., in the range of about 1075 cm -1 to about 1100 cm -1 ; e.g., in the range of about 1075 cm -1 to about 1085 cm -1 ).

123. The method according to claim 122, wherein step (b) includes determining a value of a nucleic acid (e.g., DNA, RNA, etc.) content metric that is at least partially based on the antisymmetric - PO4 and / or symmetric - PO4 absorption rate values, the nucleic acid content metric quantifying the nucleic acid content within the sample [e.g., concentration (e.g., titer)].

124. The method according to claim 123, which includes: Determining the value of the asymmetric-PO4 peak metric based on the asymmetric-PO4 absorbance value and / or determining the value of the symmetric-PO4 peak metric based on the symmetric-PO4 absorbance value; and Using the asymmetric-PO4 peak metric value and / or the symmetric-PO4 peak metric value to determine the nucleic acid content metric.

125. The method according to claim 124, wherein the antisymmetric - PO4 peak metric and / or the symmetric - PO4 peak metric are peak intensity metrics that respectively quantify the intensity [e.g., peak height, area under the curve (AUC), etc.] of the antisymmetric - PO4 band and / or the symmetric - PO4 band.

126. The method according to any one of claims 109 to 125, wherein step (b) includes determining (i) a value of a protein content metric that quantifies the protein content within the sample [e.g., concentration (e.g., titer)] and (ii) a value of a nucleic acid (e.g., DNA, RNA, etc.) content metric that quantifies the nucleic acid content within the sample [e.g., concentration (e.g., titer)], thereby independently quantifying the total protein and nucleic acid content within the sample.

127. The method according to claim 126, wherein step (b) includes determining the total capsid content as one of the virus vector sample quality metrics at least partially based on the value of the protein content metric (e.g., in the form of a function of the value of the protein content metric).

128. The method according to claim 126 or 127, wherein step (b) includes determining the full capsid fraction as one of the virus vector sample quality metrics at least partially based on (i) the value of the protein content metric and / or the total capsid content and (ii) the value of the nucleic acid content metric (e.g., in the form of a function of (i) and (ii)).

129. The method according to any one of claims 109 to 128, wherein the IR absorption rate data is or includes one or more IR absorption rate spectra, and for each specific wave number among a plurality of wave numbers spanning the measured spectral band, each IR absorption rate spectrum includes a corresponding IR absorption rate value, the corresponding IR absorption rate value representing a measure of the absorption of IR light by the aqueous sample at the specific wave number.

130. The method according to claim 129, wherein the measured spectral band spans one or more bands selected from the group consisting of: amide II band, amide III band, asymmetric - PO4 band, and symmetric - PO4 band.

131. The method according to any one of claims 109 to 130, wherein: Step (a) includes repeatedly receiving the IR absorbance data at a plurality of time points, thereby obtaining a corresponding set of IR absorbance data for each of the plurality of time points; and The method includes performing steps (b) to (c) on each set of IR absorbance data, thereby monitoring the total capsid content and / or full capsid fraction over time.

132. The method according to claim 131, which includes performing steps (a) to (c) substantially in real - time to obtain (i) a real - time capsid content signal and / or a full capsid fraction signal, the real - time capsid content signal providing a measure of the capsid content in the sample as a function of time, and the full capsid fraction signal providing a measure of the full capsid fraction in the sample as a function of time.

133. The method according to any one of claims 109 to 132, which includes: Measuring the one or more IR absorbance signals by one or more (e.g., integrated) mid-infrared (MIR) analyzers [e.g., an MIR analyzer as described in any one of claims 9 to 24].

134. The method according to claim 133, which includes measuring a corresponding one of the one or more infrared (IR) absorbance signals at each of one or more time points.

135. The method according to any one of claims 133 to 134, which includes measuring the one or more IR absorbance signals of it (the aqueous sample) when the aqueous sample flows (e.g., flows in) into the production unit and / or flows out of the production unit (e.g., the aqueous sample includes one or more input, output products, waste products, or in - process products of the production unit).

136. The method according to any one of claims 109 to 135, wherein the one or more viruses and / or virus - like particles include one or more adeno - associated viruses (AAV).

137. The method according to any one of claims 109 to 136, wherein the one or more viruses include adenovirus and / or retrovirus (e.g., lentivirus).

138. The method according to any one of claims 109 to 137, wherein the one or more viruses include plant - based viruses (e.g., tobacco mosaic virus).

139. The method according to any one of claims 109 to 138, wherein the one or more IR absorption rate signals corresponding to the IR absorption rate data are measured from the aqueous sample when the aqueous sample flows (e.g., flows in) into the production unit and / or flows out of the production unit (e.g., the aqueous sample includes one or more input, output products, waste products, or in-process products of the production unit).

140. The method according to claim 139, wherein the production unit is a purification unit.

141. The method according to claim 140, wherein the purification unit is a member selected from the group consisting of: an alternating tangential flow filtration (ATF) system, a tangential flow depth filtration (TFDF) system, a tangential flow filtration (TFF) system, a chromatography column, a direct or conventional flow filtration unit, an ultrafiltration unit, and a diafiltration unit.

142. The method according to claim 141, wherein the purification unit is a chromatography column {e.g., and wherein the chromatography column is a member selected from the group consisting of: an affinity chromatography (AC) column, a hydrophobic interaction chromatography (HIC) column, an ion exchange chromatography (IEX) column, a size exclusion chromatography (SEC) column, and a mixed mode chromatography column [e.g., any combination of the foregoing chromatography columns (e.g., IEX and HIC; e.g., IEX and SEC)]}.

143. The method according to any one of claims 139 to 142, wherein the production unit is or includes a bioreactor (e.g., a seed bioreactor; e.g., a production bioreactor).

144. The method according to any one of claims 109 to 143, wherein step (c) includes causing, by the processor, generation of one or more trigger signals (e.g., voltages) and / or transmission of one or more trigger signals (e.g., voltages) to a controller unit of the production unit, at least partially based on one or more determined viral vector sample quality metrics (e.g., their values) [e.g., the value of the determined capsid content and / or the determined full capsid fraction (e.g., its value)].

145. The method according to claim 144, wherein step (c) includes causing, by the processor, generation of a trigger signal (e.g., an analog signal) having a value at least partially based on the determined viral vector quality metric [e.g., capsid content; e.g., full capsid fraction; e.g., capsid aggregation metric].

146. The method according to any one of claims 109 to 145, wherein: Measuring, at each of the one or more time points, one or more IR absorbance signals corresponding to the IR absorbance data received in step (a) from the aqueous sample as it exits the purification unit (e.g., chromatography column); and The method includes: Using the IR absorption data to determine one or both of (i) the total capsid content and (ii) the full capsid fraction; and Using the determined total capsid content and / or full capsid fraction to control the collection of the target fraction of the aqueous sample (e.g., during a specific collection window), thereby obtaining a purified sample of viral vector material.

147. The method according to any one of claims 109 to 146, wherein: Measuring, at each of the one or more time points, one or more IR absorbance signals corresponding to the IR absorbance data received in step (a) from the aqueous sample as it exits the purification unit (e.g., chromatography column); And The method includes: Use the IR absorption data to determine a capsid aggregation metric that measures the level of aggregation between capsids in the viral vector sample; and Use the capsid aggregation metric to control the collection of the target fraction of the aqueous sample (e.g., during a specific collection window) to obtain a purified sample of viral vector material.

148. The method according to any one of claims 109 to 147, wherein the IR absorption rate data includes an IR absorption rate spectrum, and wherein step (b) includes: Receive (e.g., and / or access) one or more reference spectra, each reference spectrum measured from a corresponding (e.g., high-quality) reference sample comprising the target viral vector species and / or one or more of its model components at high purity and / or concentration (e.g., model protein solution; e.g., model ssDNA solution); and Use the IR absorption rate spectrum and the one or more reference spectra (e.g., automatically) to determine the value of at least a portion of the one or more viral vector sample quality metrics [e.g., determine one or more (e.g., multiple) measures of the deviation between the reference spectrum and the IR absorption rate spectrum as the value of the portion of the one or more viral vector sample quality metrics].

149. The method according to claim 148, wherein the one or more reference spectra include high-quality virus vector spectra measured from a reference sample having a full capsid fraction equal to or higher than a specific threshold fraction (e.g., known a priori; e.g., determined to be).

150. The method according to claim 149, wherein the threshold fraction is about 75% [e.g., about 80% (e.g., about 90%)].

151. The method according to any one of claims 148 to 150, wherein step (b) includes calculating a difference spectrum based on at least one of the one or more reference spectra and the IR absorption rate spectrum (e.g., by subtracting the IR absorption rate spectrum and / or its scaled or otherwise preprocessed version from the reference spectrum and / or its scaled or otherwise preprocessed version, or vice versa).

152. The method according to any one of claims 148 to 151, wherein step (b) includes calculating one or more derivative spectra of at least one of the one or more reference spectra and / or the IR absorption rate spectrum (e.g., first derivative; e.g., second derivative).

153. The method according to any one of claims 148 to 152, wherein step (b) includes calculating (e.g., as a measure of deviation) one or more members selected from the group consisting of: A correlation value based on the correlation between (i) a specific one of the one or more reference spectra and / or one or more of its derivatives and (ii) the IR absorption rate spectrum and / or one or more of its derivatives; a covariance value based on (i) a particular one of said one or more reference spectra and / or one or more of its derivatives and (ii) the covariance of said IR absorption rate spectrum and / or one or more of its derivatives; The Pearson correlation value between (i) a particular one of the one or more reference spectra and / or one or more of its derivatives and (ii) the IR absorption rate spectrum and / or one or more of its derivatives; and Based on the overlap integral value of the overlap integral of (i) a particular one of the one or more reference spectra and / or one or more of its derivatives and (ii) the IR absorption rate spectrum and / or one or more of its derivatives.

154. The method according to any one of claims 148 to 153, wherein step (b) comprises: Determine the value of a set of one or more specific peak metrics from the IR absorption rate spectrum to obtain a set of sample peak metric values; and Determine the measure of the deviation based on the set of sample peak metric values and a set of reference peak metric values that have been determined for the one or more specific peak metrics from the one or more reference spectra.

155. The method according to any one of claims 148 to 154, comprising determining a similarity score as a measure of said deviation, said similarity score measuring the similarity between said one or more reference spectra and said IR absorption rate spectrum.

156. The method according to any one of claims 148 to 155, comprising repetitively performing steps (a)-(c) substantially in real time, whereby the deviation from said one or more reference spectra is monitored in real time.

157. A method for (e.g., real-time) evaluating and / or monitoring the quality of viral vector content in an aqueous sample comprising a target viral vector species, the method comprising: (a) A processor of a computing device (e.g., repeatedly) receives IR absorption rate data corresponding to one or more infrared (IR) absorption rate signals measured from the sample, the IR absorption rate data comprising (e.g., at least one) IR absorption rate spectrum measured from the sample and comprising a plurality of absorption rate values, each absorption rate value associated with a specific wave number; (b) The processor receives (e.g., and / or accesses) one or more reference spectra, each reference spectrum measured from a corresponding (e.g., high-quality) reference sample comprising the target viral vector species and / or one or more of its model components at high purity and / or concentration (e.g., model protein solution; e.g., model ssDNA solution); (c) The processor uses the IR absorption rate spectrum and the one or more reference spectra (e.g., automatically) to determine one or more (e.g., multiple) measures of the deviation; (d) Stores and / or provides the measure of the deviation for display and / or further processing.

158. A system for (e.g., real-time) quantifying and / or monitoring the quality of viral vectors in an aqueous sample comprising one or more viruses and / or virus-like particles, the system comprising: A processor of a computing device; and A memory having instructions stored thereon, wherein the instructions, when executed by the processor, cause the processor to perform the method according to any one of claims 109 to 157.

159. A method for (e.g., real-time) monitoring compositional changes in a sample by infrared (IR) absorption spectroscopy, the method comprising: (a) The processor of the computing device repeatedly receives IR absorption rate data corresponding to IR absorption rate signals measured at each of a plurality of time points. For each specific time point among the plurality of time points, the IR absorption rate data includes a corresponding IR absorption rate spectrum measured from the sample at the specific time point and includes a plurality of absorption rate values, each absorption rate value being associated with a specific wave number; (b) For each specific time point among the plurality of time points, the processor (e.g., automatically) analyzes the IR absorption rate data to obtain (e.g., in real time) a normalized spectral difference signal that measures the change in normalized spectral absorption rate between consecutive time points by: Normalizing the current IR absorption rate spectrum corresponding to the specific time point using a reference absorption rate value determined from the values of the current IR absorption rate spectrum at one or more reference wave numbers to obtain a current normalized spectrum; Determining a current value of a spectral difference metric based on (e.g., calculated as) the difference between the current normalized spectrum and a previous normalized spectrum, where the previous normalized spectrum is based on one or more previously obtained IR absorption spectra (e.g., is a specific previously obtained IR absorption spectrum; e.g., is the average of a plurality of previously obtained IR absorption spectra), each previously obtained IR absorption spectrum corresponding to a specific previous time point (e.g., a specific time interval and / or its multiples before the current specific time point) and having been measured at the specific previous time point, and each specific previously obtained IR absorption spectrum having been normalized using a reference value determined from the values of the specific previously obtained IR absorption rate spectrum at the one or more reference wave numbers; And Updating the real-time normalized spectral difference signal according to the current value of the normalized spectral difference metric; And (c) The processor stores and / or provides the real-time normalized spectral difference signal for one or more of the following purposes: (i) further processing, (ii) display, and (iii) use as a control signal for adjusting one or more process parameters of a production unit (e.g., a chromatography unit; e.g., a filtration unit).

160. The method according to claim 159, further comprising (e.g., by the processor) identifying a compositional change of the sample (e.g., at a specific time) based on the real-time normalized spectral difference signal.

161. The method according to claim 160, comprising detecting a change point in the real-time normalized spectral difference signal [e.g., a change in statistical characteristics (e.g., mean, median, mode, variance, etc.); e.g., a step change], and identifying a change in the composition of the sample based on the detected change point.

162. The method according to claim 160 or 161, comprising determining (e.g., at each time point, e.g., in real-time) the value of one or more statistical parameters of the real-time normalized spectral difference signal (e.g., as a measure of sample quality).

163. The method according to claim 162, wherein the one or more statistical parameters comprise one or more members selected from the group consisting of: A mean [e.g., a running (e.g., backward-looking) mean, e.g., calculated as the mean of the real-time normalized spectral difference signal over a time window including (e.g., ending at) the current time point and one or more previous time points]; Variance [e.g., running (e.g., retrospective) variance, e.g., calculated as the variance of the real-time normalized spectral difference signal over a time window including (e.g., ending at) the current time point and one or more previous time points]; Mode; And Standard deviation.

164. The method according to claim 162 or 163, comprising identifying a compositional change based on at least one of the values of the one or more statistical parameters (i) exceeding one or more thresholds and / or (ii) changing outside a specific range [e.g., a predetermined threshold and / or range; e.g., during operation, e.g., during an initial stage of a process run (e.g., during an initial time window of a chromatographic run, e.g., during an initial ramp of a salt gradient, e.g., before protein elution) determined threshold and / or range].

165. The method according to any one of the preceding claims, wherein the sample is or comprises an aqueous sample.

166. The method according to any one of the preceding claims, wherein the sample comprises one or more protein species [e.g., a target protein species, such as a monoclonal antibody; e.g., as claimed in any one of claims 46 to 49].

167. The method according to any one of claims 166, further comprising (e.g., by the processor) identifying a compositional change (e.g., at a specific time) of the sample, the change corresponding to a change in the purity [e.g., the presence of protein species other than the target; e.g., the presence of an undesired form (e.g., non-monomeric) of the target protein species] and / or properties (e.g., secondary structure composition) of the target protein species.

168. The method according to claim 167, wherein the identified compositional change is or comprises (e.g., indicates) a change in the level and / or presence of protein aggregation within the aqueous sample.

169. The method according to claim 167 or 168, wherein the identified compositional change is or comprises (e.g., indicates) one or more members selected from the group consisting of: a change in the content (e.g., relative content) of one or more specific protein species within the (e.g., aqueous) sample; a change in the level and / or type of molecular conjugation (e.g., glycan, small molecule drug, polyethylene glycol, etc.); and a change in the content (e.g., α-helix content, β-sheet content, turn content, disordered content) of one or more protein secondary structure motifs.

170. The method according to any one of the preceding claims, wherein the sample comprises nucleic acid (e.g., DNA, RNA, mRNA, etc.).

171. The method according to claim 170, further comprising (e.g., by the processor) identifying a compositional change (e.g., at a specific time) of the (e.g., aqueous) sample, the change corresponding to a change in the purity and / or properties of the nucleic acid within the sample (e.g., a change in the relative content of one or more specific types of nucleic acid (e.g., DNA, RNA, ssDNA, dsDNA)).

172. The method according to any one of the preceding claims, wherein the aqueous sample comprises one or more viruses and / or virus-like particles [e.g., adeno-associated virus vector (AAV); e.g., lentiviral vector].

173. The method according to claim 172, further comprising (e.g., by the processor) identifying a compositional change (e.g., at a specific time) of the sample, the change corresponding to a change in the purity and / or properties of the virus and / or virus-like particle within the sample.

174. The method according to claim 173, wherein the compositional change corresponds to (e.g., indicates) a change in the relative fraction (e.g., percentage, ratio, etc. of full virus vector) of empty virus vector and / or full virus vector within the aqueous sample.

175. The method according to claim 173 or claim 174, wherein the compositional change corresponds to (e.g., indicates) the level of capsid aggregation in the sample.

176. The method according to any one of claims 173 to 175, wherein the compositional change corresponds to (e.g., indicates) a change in the relative content between viral nucleic acid from host cell proteins and host cell nucleic acid content.

177. The method according to any one of claims 160 to 176, which comprises causing (e.g., triggering) an adjustment of one or more process parameters of a production unit based on (e.g., triggered by) the identification of a compositional change in the sample (e.g., detection of a change point; e.g., based on the value of one or more statistical parameters of the real-time normalized spectral difference signal).

178. The method according to claim 177, wherein the production unit is or comprises a purification unit.

179. The method according to claim 178, wherein the purification unit is or comprises a chromatography column.

180. The method according to any one of claims 177 to 179, wherein the production unit is or comprises one or more members selected from the group consisting of: a flow controller, a valve controller (e.g., for adjusting buffer composition (e.g., by valve switching)), a temperature controller (e.g., for adjusting one or more temperature set points and / or (e.g., time) profiles).

181. The method according to any one of claims 177 to 180, which comprises triggering a response of the production unit (e.g., any parameter discussed herein) [e.g., wherein the production unit is a first production unit and the sample is associated with a second (e.g., upstream or downstream) production unit (e.g., is an input, output, or component processed by the second production unit)].

182. The method according to claims 178 to 181, wherein the sample is an aqueous sample and the method comprises causing an adjustment of a collection window to control the collection of a target fraction (e.g., monomeric species of a protein) of the aqueous sample, thereby obtaining a purified sample.

183. The method according to any one of claims 178 to 182, wherein the production unit is or comprises a filtration unit (e.g., an ultrafiltration and / or diafiltration unit) (e.g., and wherein the method comprises causing an adjustment of flow rate, transmembrane pressure, treatment time, etc., to control the composition of the retentate and / or permeate).

184. The method according to any one of claims 177 to 183, wherein the method comprises monitoring the progress of a chemical reaction based on the real-time normalized spectral difference signal (e.g., identifying compositional changes using the real-time normalized spectral difference signal) [e.g., within the production unit (e.g., bioreactor, transfection unit, pegylation unit, antibody-drug conjugation unit)].

185. The method according to any one of claims 177 to 184, wherein the method comprises causing regulation of one or more members selected from the group consisting of: in-line buffer preparation, mixing process (e.g., in a mixing tank), temperature controller.

186. The method according to any one of the preceding claims, wherein the reference absorption rate value is determined by the value of the current IR absorption rate spectrum at a single reference wave number, and the previous reference value is determined by the value of the previous IR absorption rate spectrum at the single reference wave number.

187. The method according to any one of the preceding claims, wherein determining the current value of the spectral difference metric comprises calculating the integrated absorption rate on one or more specific spectral bands (e.g., and subtracting the integrated absorption rate values) for each of the current normalized spectrum and the previous normalized spectrum.

188. The method according to claim 187, wherein the one or more specific spectral bands comprise one or more members selected from the group consisting of: Amide I spectral band (e.g., in the range of about 1600 cm -1 to about 1700 cm -1 or about 1800 cm -1 (e.g., in the range of about 1600 cm -1 to about 1725 cm -1 ; e.g., in the range of about 1625 cm -1 to about 1725 cm -1 ; e.g., in the range of about 1630 cm -1 to about 1650 cm -1 ), and / or the amide II region, which ranges from about 1500 to about 1600 cm -1 (e.g., in the range of about 1500 cm -1 to about 1575 cm -1 ; e.g., in the range of about 1500 cm -1 to about 1550 cm -1 ; e.g., in the range of about 1540 cm -1 to about 1560 cm -1 )); Amide II spectral band (e.g., in the range of about 1500 cm -1to about 1600 cm -1 e.g., in the range of about 1500 cm -1 to about 1575 cm -1 e.g., in the range of about 1500 cm -1 to about 1550 cm -1 e.g., in the range of about 1540 cm -1 to about 1560 cm -1 ); and amide III spectral band (e.g., in the range of about 1250 cm -1 to about 1350 cm -1 e.g., in the range of about 1250 cm -1 to about 1325 cm -1 ; e.g., in the range of about 1275 cm -1 to about 1325 cm -1 ; e.g., in the range of about 1280 cm -1 to about 1300 cm -1 ).

189. The method according to claim 187 or claim 188, wherein the one or more specific spectral bands include one or more members selected from the group consisting of: asymmetric - PO4 spectral band (e.g., in the range of about 1150 cm -1 to about 1250 cm -1 e.g., in the range of about 1175 cm -1 to about 1250 cm -1 ; e.g., in the range of about 1200 cm -1 to about 1250 cm -1 ; e.g., in the range of about 1210 cm -1 to about 1230 cm -1 ); and symmetric - PO4 spectral band (e.g., in the range of about 1000 cm -1 to about 1100 cm -1 e.g., in the range of about 1050 cm -1 to about 1100 cm -1 ; e.g., in the range of about 1075 cm -1 to about 1100 cm -1 ; e.g., in the range of about 1075 cm -1 to about 1085 cm -1 ).

190. The method according to any one of the preceding claims, further comprising measuring the IR absorption rate signal using one or more MIR analyzers (e.g., as described in any one of claims 9 to 24).

191. A method for monitoring temporal changes of a sample by infrared (IR) absorption spectroscopy (e.g., in real time), the method comprising: (a) The processor of the computing device repeatedly receives IR absorption rate data corresponding to IR absorption rate signals measured at each of a plurality of time points. For each specific time point among the plurality of time points, the IR absorption rate data includes a corresponding IR absorption rate spectrum measured from the sample at the specific time point and includes a plurality of absorption rate values, each absorption rate value being associated with a specific wave number; (b) The processor (e.g., automatically) analyzes the IR absorption rate data to obtain one or both of the following: (e.g., in real-time) a time-differential signal that measures a change over time in values of one or more features (e.g., a sample quality metric) determined using (i) a first set of one or more IR absorption rate spectra corresponding to a first set of specific time points and (ii) a second set of one or more IR absorption rate spectra corresponding to the first set of specific time points; and (e.g., in real-time) a time-aggregation signal that is a function (e.g., a running sum, average, median, mode, variance, standard deviation, etc. over a specific time window) of at least a portion of the plurality of time points [e.g., a cumulative increasing portion, e.g., starting at a specific time point and ending at the current time point; e.g., a time window of a specific size (e.g., a look-back window)]; and (c) the processor of the computing device stores and / or provides the time-differential signal and / or the time-aggregation signal for one or more of the following uses: (i) further processing, (ii) display, and (iii) use as a control signal for adjusting one or more process parameters of a production unit (e.g., a chromatography unit; e.g., a filtration unit).

192. A system for monitoring temporal (e.g., compositional) changes of a sample by infrared (IR) absorption spectroscopy (e.g., in real time), the system comprising: a processor of a computing device; and a memory having instructions stored thereon, wherein the instructions, when executed by the processor, cause the processor to perform the method according to any one of claims 159 to 191.

193. The system according to claim 192, further comprising one or more MIR analyzers (e.g., as described in any one of claims 9 to 24).

194. The system according to claim 192 or 193, comprising a production unit (e.g., a chromatography unit; e.g., a filtration unit).

195. A method for obtaining a purified sample of a target protein species by bioprocess monitoring and control based on mid-infrared (IR) spectroscopy, the method comprising: (a) at each of a plurality of time points, measuring, by one or more mid-infrared (MIR) analyzers, corresponding mid-infrared absorption rate spectra of an aqueous sample exiting a purification unit, the aqueous sample including one or more protein species that include the target protein species, thereby measuring a plurality of mid-infrared absorption rate spectra over time; (b) receiving, by a processor of a computing device, spectral data corresponding to the measured plurality of mid-infrared absorption rate spectra; (c) for each of at least a portion of the plurality of time points, determining, by the processor based on the spectral data, corresponding values of one or more sample quality metrics, the one or more sample quality metrics including a measure of the concentration and / or purity of the target protein species in the aqueous sample; and (d) using the determined values of the one or more sample quality metrics to control the collection of the target fraction of the aqueous sample (e.g., during a specific collection window), thereby obtaining the purified sample of the target protein species.

196. The method according to claim 195, wherein the purification unit is or comprises a chromatography column {e.g., and wherein the chromatography column is a member of the group consisting of: an affinity chromatography (AC) column, a hydrophobic interaction chromatography (HIC) column, an ion exchange chromatography (IEX) column, a size exclusion chromatography (SEC) column, and a mixed mode chromatography column [e.g., any combination of the foregoing chromatography columns (e.g., IEX and HIC; e.g., IEX and SEC)]}.

197. The method according to claim 195 or 196, wherein the purification unit is or comprises an ultrafiltration and diafiltration system (UF / DF) [e.g., a tangential flow filtration (TFF) system].

198. The method according to any one of claims 195 to 197, wherein the target protein species is selected from the group consisting of: monoclonal antibody (mAb), fusion protein, viral capsid protein, antibody-drug conjugate, recombinant protein, and plasma protein.

199. The method according to any one of claims 195 to 198, wherein the aqueous sample comprises multiple different molecular forms of a specific protein [e.g., a therapeutic protein (e.g., mAb)], including monomeric forms and one or more aggregated forms (e.g., dimers and / or other multimers), and wherein the target protein species is the monomeric form of the specific protein.

200. The method according to any one of claims 195 to 199, wherein the aqueous sample comprises one or more subspecies of a specific protein, each subspecies having a specific desired level and / or type of molecular conjugation (e.g., glycan, small molecule drug, polyethylene glycol, etc.), and wherein the target protein species is a specific one of the one or more subspecies.

201. The method according to any one of claims 195 to 200, wherein the one or more MIR analyzers comprise a mid-infrared spectrometer based on a quantum cascade laser (QCL), which comprises: A QCL-based source that is aligned and operable to emit a MIR beam [e.g., including one or more wavelengths substantially within the MIR spectral range (e.g., in the range of about 5000 cm -1 to about 500 cm -1 (e.g., about 2 to 20 microns))]; one or more sampling optics that are aligned to direct and / or allow the MIR beam and / or at least a portion thereof to pass through and / or contact at least a portion of the aqueous sample [e.g., wherein the MIR beam contacts the portion of the aqueous sample by reflection at an interface between a solid material (e.g., an ATR crystal and / or an optical fiber) and the aqueous sample (e.g., wherein the MIR beam undergoes total internal reflection and contacts / detects the portion of the aqueous sample through an evanescent wave extending into the aqueous sample)], and, after passing through or contacting the portion of the aqueous sample, towards one or more detectors; and The one or more detectors are aligned and operable to detect the MIR beam after the MIR beam has passed through and / or contacted the aqueous sample.

202. The method according to claim 201, wherein: The one or more sampling optics include a flow cell that includes a detection channel through which the aqueous sample flows; and The one or more detectors are aligned and operable to detect the MIR beam exiting the detection channel after the MIR beam has transmitted through the detection channel.

203. The method according to claim 201 or 202, wherein the QCL-based source is a tunable QCL capable of operating to scan the emission frequency of the MIR beam through multiple frequencies within a scan range (e.g., wherein the scan range includes a range from about 1700 cm -1 to 1400 cm -1 ; e.g., wherein the scan range includes a range from 1300 cm -1 to 1050 cm -1 ; e.g., wherein the scan range includes a range of at least 1200 cm -1 to 1000 cm -1 ), and the method comprises, at each of the one or more time points: scanning the emission frequency of the MIR beam within the scan range of the tunable laser, thereby irradiating the aqueous sample at multiple emission frequencies; and Detect the MIR beam at each of the plurality of emission frequencies using the one or more detectors (e.g., which has (i) been internally reflected by the interface between the high refractive index material and the aqueous sample and / or (ii) transmitted through the detection channel through which the aqueous sample flows), thereby measuring a corresponding infrared (IR) spectrum including a plurality of values as the corresponding IR absorption rate signal of the aqueous sample, each value of the plurality of values being associated with and representative of and / or based on the power detected at a particular one of the plurality of emission frequencies).

204. The method according to any one of claims 195 to 203, wherein the MIR analyzer is an in-line sensor (e.g., as opposed to an off-line or at-line sensor), and wherein step (a) comprises repeatedly measuring the IR absorption rate spectrum over time {e.g., every 20 seconds or less [e.g., every 10 seconds or less (e.g., every 5 seconds or less; (e.g., every second or less))]} as the aqueous solution exits the purification unit (e.g., thereby measuring the IR absorption rate spectrum of the aqueous sample substantially in real time).

205. The method according to any one of claims 195 to 204, wherein for each of the one or more time points, the spectral data includes a corresponding amide band spectrum [e.g., for each particular wavelength of a plurality of sampling (e.g., emission) wavelengths in the range of about 1800 to about 800 cm -1 in the range (e.g., in the range of about 1700 to 1400 cm -1 in the range), the amide band spectrum including a relevant absorption value representing the absorption level of a portion of the aqueous sample at the particular wavelength].

206. The method according to any one of claims 195 to 205, wherein step (c) comprises: The reference spectrum of the target protein species is received (e.g., and / or accessed) by the processor, the reference spectrum having been measured from a specific corresponding reference sample that includes the target protein species in substantially isolated and / or high purity [e.g., 75% purity or better (e.g., 90% purity or better (e.g., 95% purity or better))]; and The reference spectrum is repeatedly used at each of the plurality of time points to determine the concentration of the target protein species in the aqueous sample at each time point, thereby tracking the concentration of the target protein species over time.

207. The method according to any one of claims 195 to 206, wherein step (c) comprises: One or more impurity reference spectra are received (e.g., and / or accessed) by the processor, each impurity reference spectrum being associated with a specific impurity of interest and having been measured from a specific corresponding reference sample that includes the impurity of interest in substantially isolated and / or high purity [e.g., 75% purity or better (e.g., 90% purity or better (e.g., 95% purity or better))]; and The one or more impurity reference spectra are repeatedly used at each of the plurality of time points to determine the concentration of each of the impurities of interest in the aqueous sample.

208. The method according to any one of claims 195 to 207, wherein for each of the one or more time points, the spectral data includes a corresponding amide band spectrum, and wherein step (c) comprises determining the absorption rate ratio at at least two wavenumbers within the amide band spectrum as a measure of sample purity.

209. The method according to any one of claims 195 to 208, wherein step (d) comprises causing the processor to transmit one or more trigger signals (e.g., voltage) to the controller unit of the purification unit and / or a downstream (downstream of the purification unit) valve.

210. The method according to claim 209, wherein step (d) comprises one or both of the following: initiating, by the controller unit, collection of the target fraction of the aqueous sample based on the one or more trigger signals [e.g., wherein a particular one of the one or more trigger signals is an analog signal, and the controller unit initiates collection of the target fraction based on the amplitude of the analog signal (e.g., when it exceeds a particular threshold; e.g., when it drops below a particular threshold); e.g., wherein a particular one of the one or more trigger signals is a digital signal that triggers (e.g., by transitioning from a 0 voltage level to a 1 voltage level, or vice versa) the initiation of collection of the target fraction], and stopping, by the controller unit, collection of the target fraction of the aqueous sample based on the one or more trigger signals [e.g., wherein a particular one of the one or more trigger signals is an analog signal, and the controller unit stops collection of the target fraction based on the amplitude of the analog signal (e.g., when it exceeds a particular threshold; e.g., when it drops below a particular threshold); e.g., wherein a particular one of the one or more trigger signals is a digital signal that triggers (e.g., by transitioning from a 0 voltage level to a 1 voltage level, or vice versa) the stopping of collection of the target fraction].

211. The method according to any one of claims 195 to 210, wherein the target protein species is the monomeric form of a specific protein (e.g., a monoclonal antibody), and the method comprises: In step (c), the values of the following are determined over time: (i) the concentration of the monomeric form of the specific protein and / or (ii) the cumulative purity of the monomeric form of the specific protein in the total collected volume of the sample exiting the purification unit [e.g., the relative fraction (e.g., mass) of the monomeric form of the specific protein collected relative to the total protein collected]; and In step (d), collection of the aqueous sample exiting the purification unit is stopped at a specific stop time, at least in part based on the values of the concentration and / or cumulative purity of the monomeric form of the specific protein.

212. The method according to claim 211, wherein the aqueous sample comprises (i) one or more highly aggregated forms of the specific protein and / or (ii) one or more fragmented species of the specific protein, and wherein step (c) comprises determining, over time, the concentration of the one or more aggregated forms of the specific protein and / or the concentration of the one or more fragmented species of the specific protein.

213. The method according to any one of claims 195 to 212, wherein the aqueous sample comprises one or more excipients, and the method comprises, in step (c): Determine values of the concentration and / or amount of the one or more excipients in the aqueous sample exiting the purification unit at one or more time points based on the spectral data; and In step (d), use the determined values of the excipient concentration and / or amount to control the collection of the target fraction of the aqueous sample.

214. A method for preparing a biopharmaceutical formulation comprising one or more excipients, the method comprising: (a) Receiving a solution comprising a purified active pharmaceutical ingredient that includes a protein species (e.g., a monoclonal antibody); (b) Injecting and / or mixing one or more excipients into the solution of the purified active pharmaceutical ingredient over a period of time, thereby producing an in-process active pharmaceutical ingredient solution that includes the purified active pharmaceutical ingredient and the one or more excipients, wherein the relative concentrations of the purified active pharmaceutical ingredient and the one or more excipients vary over the period of time when the one or more excipients are injected and / or mixed; (c) Measuring, by one or more mid-infrared (MIR) analyzers, one or both of the following at each of one or more time points: (i) The corresponding mid-infrared absorption rate spectrum from the in-process active pharmaceutical ingredient solution; and (ii) The corresponding mid-infrared absorption rate spectrum from a stock solution comprising at least one of the one or more excipients, thereby measuring one or more mid-infrared absorption rate spectra; (d) The spectral data corresponding to the measured one or more mid-infrared absorption rate spectra is received by a processor of the computing device; (e) For each of at least a portion of the one or more time points, the processor determines corresponding values of one or more sample quality metrics based on the spectral data, the one or more sample quality metrics including (i) the protein species and / or (ii) one or more measures of the concentration and / or purity of a subset of the one or more excipients; and (f) The determined values of the one or more sample quality metrics are used to control the injection and / or mixing of the one or more excipients, thereby obtaining a final drug substance having a desired protein and / or excipient content and / or purity.

215. The method according to claim 214, wherein step (b) comprises using an ultrafiltration / diafiltration (UF / DF) system (e.g., for buffer exchange).

216. The method according to claim 214 or 215, wherein the protein species is or comprises a monoclonal antibody.

217. The method according to any one of claims 214 to 216, wherein the one or more excipients are or comprise one or more surfactants {e.g., detergents; e.g., wetting agents and / or solubilizers [e.g., polysorbate 20 (Tween 20), polysorbate 80 (Tween 80), poloxamer (Pluronic F68 and F127), Triton X-100, Brij 30, Brij 35, etc.]}.

218. The method according to any one of claims 214 to 217, wherein the one or more excipients are or comprise one or more fillers {e.g., sugars and / or polyols [e.g., sucrose, trehalose, glucose, lactose, sorbitol, mannitol, glycerol, etc.]; e.g., amino acids [e.g., arginine, aspartic acid, glutamic acid, lysine, proline, glycine, histidine, methionine, alanine, etc.]; e.g., polymers and proteins [e.g., gelatin, PVP, PLGA, PEG, dextran, cyclodextrin and derivatives, starch derivatives, HSA, BSA]}.

219. The method according to any one of claims 214 to 218, wherein the one or more MIR analyzers comprise a mid-infrared spectrometer based on a quantum cascade laser (QCL), which comprises: A QCL-based source that is aligned and operable to emit a MIR beam [e.g., including one or more wavelengths that are substantially within the MIR spectral range (e.g., in the range of about 5000 cm -1 to about 500 cm -1 (e.g., about 2 to 20 microns))]; One or more sampling optical devices, which are aligned to direct and / or allow the MIR beam and / or at least a portion thereof to pass through and / or contact at least a portion of the stock solution and / or a portion of the in-process drug substance solution [e.g., wherein the MIR beam contacts the at least a portion of the stock solution and / or the portion of the in-process drug substance solution by reflection at an interface between a solid material (e.g., an ATR crystal and / or an optical fiber) and the at least a portion of the stock solution and / or the portion of the in-process drug substance solution (e.g., wherein the MIR beam undergoes total internal reflection and contacts / detects the at least a portion of the stock solution and / or the portion of the in-process drug substance solution by an evanescent wave extending into the aqueous sample)], and after passing through or contacting the at least a portion of the stock solution and / or the portion of the in-process drug substance solution, towards one or more detectors; and The one or more detectors, which are aligned and operable to detect the MIR beam after the MIR beam passes through and / or contacts the at least a portion of the stock solution and / or the portion of the in-process drug substance solution.

220. The method according to claim 219, wherein: The one or more sampling optical devices include a flow cell, the flow cell including a detection channel through which the at least a portion of the stock solution and / or the portion of the in-process drug substance solution flows; and The one or more detectors are aligned and operable to detect the MIR beam exiting the detection channel after the MIR beam transmits through the detection channel.

221. The method according to claim 219 or 220, wherein the QCL-based source is a tunable QCL capable of operating to scan the emission frequency of the MIR beam through a plurality of frequencies within a scan range (e.g., wherein the scan range includes a range from about 1700 cm -1 to 1400 cm -1 ; e.g., wherein the scan range includes a range from 1300 cm -1 to 1050 cm -1 ; e.g., wherein the scan range includes a range from at least 1200 cm -1 to 1000 cm -1 ), and the method comprises, at each of the one or more time points: scanning the emission frequency of the MIR beam within the scan range of the tunable laser, thereby irradiating the portion of the stock solution and / or the portion of the in-process API solution at a plurality of emission frequencies; and detecting the MIR beam at each of the plurality of emission frequencies with the one or more detectors (e.g., having (i) been internally reflected by the interface between the high refractive index material and the portion of the stock solution and / or the portion of the in-process API solution and / or (ii) transmitted through the detection channel through which the portion of the stock solution and / or the portion of the in-process API solution has flowed), thereby measuring a corresponding infrared (IR) spectrum comprising a plurality of values as a corresponding IR absorption rate signal of the portion of the stock solution and / or the portion of the in-process API solution, each value being associated with and representing and / or based on the power detected at a particular one of the plurality of emission frequencies.

222. The method according to any one of claims 214 to 221, wherein the MIR analyzer is an on-line sensor (e.g., as opposed to an off-line or at-line sensor), and wherein step (c) comprises repeatedly measuring the IR absorption rate spectrum over time (e.g., every 20 seconds or less [e.g., every 10 seconds or less (e.g., every 5 seconds or less; (e.g., every second or less))]) when one or more excipients are being injected and / or mixed (e.g., so as to measure the IR absorption rate spectrum of the API solution in the process substantially in real time).

223. The method according to any one of claims 214 to 222, wherein for each of the one or more time points, the spectral data comprises corresponding amide band spectra [e.g., for each specific wavelength of a plurality of sampled (e.g., emitted) wavelengths in the range of about 1800 to about 800 cm -1 (e.g., in the range of about 1700 to 1400 cm -1 ), the amide band spectra comprising associated absorption values representing the absorption level of a portion of the aqueous sample at the specific wavelength].

224. The method according to any one of claims 214 to 223, wherein for each of the one or more time points, the spectral data comprises corresponding sugar band spectra [e.g., for each specific wavelength of a plurality of sampled (e.g., emitted) wavelengths in the range of about 1400 to about 800 cm -1 (e.g., in the range of about 1200 to 1000 cm -1 ), the sugar band spectra comprising associated absorption values representing the absorption level of a portion of the aqueous sample at the specific wavelength].

225. The method according to any one of claims 214 to 224, wherein step (e) comprises: The reference spectrum of the protein species is received (e.g., and / or accessed) by the processor, the reference spectrum having been measured from a specific corresponding reference sample comprising the protein species that is substantially separated and / or of high purity [e.g., 75% purity or better (e.g., 90% purity or better (e.g., 95% purity or better))]; and The reference spectrum is repeatedly used at each of the plurality of time points to determine the concentration of the protein species in the in-process drug substance solution at each time point, thereby tracking the concentration of the protein species over time.

226. The method according to any one of claims 214 to 225, wherein step (e) comprises: Receive (e.g., and / or access) by the processor one or more excipient reference spectra, each excipient reference spectrum being associated with a particular excipient of interest (among the one or more excipients) and having been measured from a particular corresponding reference sample comprising the particular excipient of interest that is substantially separated and / or of high purity [e.g., 75% purity or better (e.g., 90% purity or better (e.g., 95% purity or better))]; And Repeatedly use the one or more excipient reference spectra at each of the plurality of time points to determine the concentration of each of the one or more excipients of interest in the stock solution and / or the in-process API solution.

227. The method according to any one of claims 214 to 226, wherein step (f) comprises causing, by the processor, transmission of one or more trigger signals (e.g., voltage) to a controller unit (e.g., the controller unit of a UF / DF system; e.g., the controller unit of one or more valves).

228. The method according to claim 227, wherein step (f) comprises one or both of the following: initiating, by the controller unit, injection and / or mixing of the one or more excipients [e.g., wherein a particular one of the one or more trigger signals is an analog signal and the controller unit initiates the injection and / or mixing based on the amplitude of the analog signal (e.g., when it exceeds a particular threshold; e.g., when it drops below a particular threshold); e.g., wherein a particular one of the one or more trigger signals is a digital signal that triggers (e.g., by transitioning from a 0 voltage level to a 1 voltage level, or vice versa) the injection and / or mixing], and stopping, by the controller unit, injection and / or mixing of the one or more excipients based on the one or more trigger signals [e.g., wherein a particular one of the one or more trigger signals is an analog signal and the controller unit stops the injection and / or mixing of the one or more excipients based on the amplitude of the analog signal (e.g., when it exceeds a particular threshold; e.g., when it drops below a particular threshold); e.g., wherein a particular one of the one or more trigger signals is a digital signal that triggers (e.g., by transitioning from a 0 voltage level to a 1 voltage level, or vice versa) the stopping of the injection and / or mixing of the one or more excipients].

229. A system for obtaining a purified sample of a target protein species by bioprocess monitoring and control based on mid-infrared (IR) spectroscopy, the system comprising: One or more mid-infrared (MIR) analyzers [e.g., each MIR analyzer being operable to (e.g., based on one or more signals / communicating with a processor) measure, at each of a plurality of time points, a corresponding mid-infrared absorbance spectrum of an aqueous sample exiting a purification unit, the aqueous sample comprising one or more protein species, the protein species comprising the target protein species, thereby measuring a plurality of mid-infrared absorbance spectra over time]; A processor of a computing device; And A memory having instructions stored thereon, wherein the instructions, when executed by the processor, cause the processor to: (a) Receive spectral data corresponding to a plurality of measured mid-infrared absorbance spectra, each mid-infrared absorbance spectrum having been measured by the one or more MIR analyzers at a corresponding one of the plurality of time points from an aqueous sample exiting a purification unit, the aqueous sample comprising one or more protein species, the protein species comprising the target protein species; (b) For each of at least a portion of the plurality of time points, determine a corresponding value of one or more sample quality metrics based on the spectral data, the one or more sample quality metrics including a measure of the concentration and / or purity of the target protein species in the aqueous sample; And (c) Use the determined values of the one or more sample quality metrics to control the collection of a target fraction of the aqueous sample (e.g., during a specific collection window), thereby obtaining the purified sample of the target protein species.

230. A system for preparing a biopharmaceutical formulation comprising one or more excipients, the system comprising: One or more mid-infrared (MIR) analyzers; A processor of a computing device; And A memory having instructions stored thereon, wherein the instructions, when executed by the processor, cause the processor to: (a) Receive spectral data corresponding to one or more mid-infrared absorbance spectra, the mid-infrared absorbance spectra having been measured by the one or more MIR analyzers at each of one or more time points from one or both of the following: (i) An in-process API solution that includes a purified API and one or more excipients injected into and / or mixed with it over time; And (ii) A stock solution that includes at least one of the one or more excipients; (b) For each of at least a portion of the one or more time points, determine corresponding values of one or more sample quality metrics based on the spectral data, the one or more sample quality metrics including (i) the protein species and / or (ii) one or more measures of the concentration and / or purity of a subset of the one or more excipients; and (c) Use the determined values of the one or more sample quality metrics to control the injection and / or mixing of the one or more excipients, thereby obtaining a final drug substance having a desired protein and / or excipient content and / or purity.

Images