Systems and methods for monitoring and controlling bioproduction processes using mid-infrared spectroscopy

Abstract

Presented herein are methods and systems for monitoring and / or controlling production units used at various stages in the development and manufacture of biological products. In particular, in certain embodiments, the bioproduction monitoring and control techniques described herein utilize mid-infrared (mid-IR) analyzers capable of obtaining mid-IR spectral data from aqueous samples in substantially 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, and aggregation, as well as viral and / or nucleic acid properties.

Description

[Technical Field]

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS This application claims priority to and the benefit of U.S. Provisional Application No. 63 / 425,504, filed November 15, 2022, U.S. Provisional Application No. 63 / 431,989, filed December 12, 2022, and U.S. Provisional Application No. 63 / 438,969, filed January 13, 2023, the contents of each of which are hereby incorporated by reference in their entirety. [Background technology]

[0002] background Biological products (also called biopharmaceuticals) are a rapidly developing and increasingly important class of products obtained (e.g., isolated) from natural sources, such as humans, animals, or microorganisms. They can include a variety 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, autoimmunity / inflammation, exposure to infectious agents, and genetic disorders. Gene therapy has immense potential to transform the lives of patients with genetic diseases.

[0003] Unlike traditional small molecule drugs, which are chemically synthesized and have known structures, biopharmaceuticals are produced through highly complex biological processes and are complex mixtures that are difficult to identify and / or characterize either as the final product or at various steps along the production line. Multi-step purification processes are typically required to obtain a consistent, pure, and effective final product.

[0004] The difficulties associated with accurately identifying and characterizing biological products and their components along the manufacturing cycle are thus an obstacle to the development and testing of new biopharmaceutical manufacturing processes, post-approval and / or expansion of production capacity to address fluctuations in demand, and the refinement and control of current procedures to ensure consistent product quality, avoid adverse events, and reduce downtime. Thus, improved technology for monitoring the inputs and outputs of biological product processing steps and manufacturing control is needed. Summary of the Invention [Means for solving the problem]

[0005] Abstract Presented herein are methods and systems for monitoring and / or controlling production units used at various stages in the development and manufacture of biological products. In particular, in certain embodiments, the bioproduction monitoring and control techniques described herein utilize mid-infrared (mid-IR) analyzers capable of obtaining mid-IR spectral data from aqueous samples in substantially 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, as well as viral and / or nucleic acid properties such as empty / full capsid size and, in certain embodiments, other sample properties that may serve as critical quality attributes (CQAs) according to U.S. Food and Drug Administration (FDA) guidelines for pharmaceutical development. Sample quality metrics can thus be measured substantially in real time and / or continuously to continuously assess production quality, for example, for protein therapeutics and gene therapeutics.

[0006] Control systems and methods can then use this real-time data to control and / or adjust process parameters, such as collection windows, flow rates, salt gradients, etc., for collecting target fractions during the chromatographic elution process to improve target recovery, sample purity, potency, stability, etc.

[0007] Thus, by facilitating control and improvement of downstream sample processing, the technology described herein enables more efficient and robust processing, as well as improved results in terms of product quality (e.g., purity and potency) and reduced variability, which leads to improved success rates in FDA testing, faster time to market, easier scale-up, and reduced costs, ultimately resulting in more effective and accessible therapeutics for patients in need.

[0008] In one aspect, the disclosure relates to a method for obtaining a purified sample of a target protein species through real-time monitoring of protein heterogeneity and (e.g., automated; e.g., semi-automated) control of purification processing, the method comprising: (a) measuring, by one or more mid-infrared (MIR) analyzers, at each of one or more time points, a corresponding infrared (IR) absorbance signal from 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, wherein the one or more sample quality metrics include a protein aggregation metric indicative of a level of protein aggregation within the aqueous sample; and (d) obtaining a purified sample of the target protein species by using the one or more sample quality metrics to control collection of target fractions of the aqueous sample (e.g., during a specific collection window).

[0009] In certain embodiments, the purification unit is or comprises a chromatography column, hi 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 thereof (e.g., IEX and HIC; e.g., IEX and SEC)).

[0010] In certain embodiments, the target protein species is or comprises one or more members selected from the group consisting of a monoclonal antibody (mAb), a fusion protein, a viral capsid protein, an antibody-drug conjugate, a recombinant protein, and a plasma protein. In certain embodiments, the target protein species is or comprises a peptide chain and / or a protein fragment.

[0011] In certain embodiments, the aqueous sample contains a plurality of different protein species. In certain embodiments, the aqueous sample contains a heterogeneous population of target protein species, including monomeric and aggregated portions. 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., glycans, small molecule drugs, polyethylene glycol, etc.).

[0012] In certain embodiments, at least one of the one or more MIR analyzers is configured to analyze MIR light [e.g., a wavelength range substantially within the MIR spectral range (e.g., about 5000 cm -1 ~about 500cm -1the beam of MIR light passes through and / or contacts at least a portion of the aqueous sample (e.g., the beam of MIR light 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., the beam of MIR light undergoes total internal reflection and contacts / probes the portion of the aqueous sample by evanescent waves that extend into the aqueous sample) and one or more sampling optics aligned to direct and / or allow the passage of the beam of MIR light, and / or at least a portion thereof, towards one or more detectors after passage through or contact with the portion of the aqueous sample; and one or more detectors aligned and operable to detect the beam of MIR light after its passage through and / or contact with the aqueous sample.

[0013] In certain embodiments, the one or more sampling optics include a high refractive index material (e.g., an ATR crystal; e.g., an optical fiber) aligned such that the beam of MIR light is directed toward, incident on, and internally reflected by (e.g., back into) an interface between the high refractive index material and the aqueous sample (e.g., such that the beam of MIR light is incident on the interface at an angle greater than the critical angle for total internal reflection); and one or more detectors aligned and operable to detect the beam of MIR light emerging from the high refractive index material after its internal reflection 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.

[0014] In certain embodiments, the one or more sampling optics include a flow cell including a detection channel through which the aqueous sample flows; and the one or more detectors are aligned and operable to detect the beam of MIR light exiting the detection channel after its transmission through the detection channel. In certain embodiments, the path length through the detection channel (e.g., preceding the beam of MIR light upon transmission) is about 10 μm or longer (e.g., about or at least 15 μm or longer; e.g., about 25 μm or longer; e.g., about 30 μm or longer; e.g., about 40 μm or longer; e.g., about 50 μm or longer).

[0015] In certain embodiments, one or more MIR analyzers detect MIR light [e.g., a range of wavelengths substantially within the MIR spectral range (e.g., about 5000 cm -1 ~about 500cm -1 The MIR spectrometer is, or comprises, a quantum cascade laser (QCL)-based MIR spectrometer comprising a QCL light source operable to emit a beam of wavelengths ranging from about 2 to 20 μm (e.g., in the range of about 2 to 20 μm).

[0016] In certain embodiments (e.g., when the MIR light source is a laser), the beam of MIR light is about 4 cm -1 or less (e.g., about 2 cm -1 or less; for example, about 1 cm -1 or less; for example, about 0.5 cm -1 or less).

[0017] In certain embodiments, the power of the beam of MIR light 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).

[0018] In one particular embodiment, the spectral resolution of the MIR spectrometer is about 4 cm -1or better (e.g., less than) [e.g., about 2 cm -1 or better (e.g., less than); e.g., about 1 cm -1 or better (e.g., less than); e.g., about 0.5 cm -1 or better (e.g., less than); e.g., about 0.25 cm -1 or better (e.g., less than); e.g., about 0.1 cm -1 or better (e.g., less than); e.g., about 0.05 cm -1 or better (e.g., less than)].

[0019] In certain embodiments, the (e.g., frequency / wavelength) accuracy of the MIR spectrometer is approximately 2 cm -1 or better (e.g., less than) [e.g., about 1 cm -1 or better (e.g., less than); e.g., about 0.5 cm -1 or better (e.g., less than); e.g., about 0.25 cm -1 or better (e.g., less than); e.g., about 0.1 cm -1 or better (e.g., less than); e.g., about 0.01 cm -1 or better (e.g., less than)].

[0020] In certain embodiments, the (e.g., frequency / wavelength) repeatability of the MIR spectrometer is about 0.5 cm -1 or better (e.g., less than) [e.g., about 0.25 cm -1 or better (e.g., less than); e.g., about 0.1 cm -1 or better (e.g., less than); e.g., about 0.05 cm -1 or better (e.g., less than); e.g., about 0.001 cm -1 or better (e.g., less than)].

[0021] In certain embodiments, the MIR light source is a tunable laser (e.g., a tunable QCL) (e.g., operable to sweep the emission frequency of a beam of MIR light through multiple frequencies over a scan range), and the method includes illuminating the aqueous sample with multiple emission frequencies by sweeping the emission frequency of the beam MIR light over the scan range of the tunable laser at each of one or more time points; and detecting, with one or more detectors, the beam of MIR light (e.g., (i) internally reflected by an interface between the 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 multiple emission frequencies, thereby measuring a corresponding infrared (IR) spectrum including multiple values ​​associated with, representing, and / or based on the detected output at a particular one of the multiple emission frequencies as a corresponding IR absorbance signal from the aqueous sample.

[0022] In certain embodiments, the plurality of radiation wavelengths ranges from about 1800 to about 800 cm -1 (For example, about 1725 to about 1025 cm -1 For example, about 1750 to about 1350 cm -1 For example, about 1725 to about 1375 cm -1 For example, about 1700 to about 1500 cm -1 For example, about 1700 to about 1600 cm -1 For example, about 1700 to about 1000 cm -1 ) In one particular embodiment, the plurality of radiation wavelengths comprises one or more wavelengths within a spectral band ranging from about 1600 cm -1 ~Approx. 1500cm -1 It includes one or more wavelengths within a spectral band in the range of

[0023] In certain embodiments, the MIR analyzer is an online sensor (e.g., as opposed to an offline or at-line sensor) that measures the IR absorbance signal substantially in real time as the aqueous solution exits the purification unit.

[0024] In certain embodiments, the IR absorbance data is, for each of one or more time points, a corresponding amide band spectrum [e.g., an amide band spectrum from about 1800 to about 800 cm -1 for each particular wavelength of a plurality of sampled (e.g., radiation) wavelengths in the range, an associated absorption value representing the level of absorption by a portion of the aqueous sample at the particular wavelength.

[0025] In certain embodiments, step (d) comprises determining, for each particular time point of at least some of the one or more time points, a corresponding value of the protein aggregation metric.

[0026] In certain embodiments, for each particular time point, determining a corresponding value of the protein aggregation metric includes calculating, from the amide band spectrum corresponding to the particular time point, a value of an amide II peak metric that quantifies one or more properties (e.g., frequency position, linewidth, intensity) of the amide II band at the particular time point; and using the value of the amide II peak metric to determine a corresponding value of the protein aggregation metric (e.g., the protein aggregation metric is, or is a function of, the amide II peak metric).

[0027] In certain embodiments, the amide II peak metric is a frequency location metric that quantifies the frequency that is substantially centered at a particular time point for the amide II band (e.g., center of mass frequency, frequency of maximum height of the amide II band, center frequency of a fitted peak function (e.g., Gaussian, Lorentzian, etc.), etc.). In certain embodiments, the frequency location metric is the center of mass frequency of the amide II band.

[0028] In certain embodiments, for each particular time point, determining a corresponding value of the protein aggregation metric includes calculating, from the amide band spectrum corresponding to the particular time point, a value of an amide I peak metric that quantifies one or more characteristics (e.g., frequency position, linewidth, intensity) of the amide I band at the particular time point; and using both the amide I peak metric value and the amide II peak metric value to determine a corresponding value of the protein aggregation metric (e.g., the protein aggregation metric is a function of the amide I peak metric and the amide II peak metric).

[0029] In certain embodiments, the amide I peak metric is a peak intensity metric that quantifies the intensity of the amide I band at a particular time point (e.g., peak height of the amide I band, 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 a particular time point (e.g., peak height of the amide I band, area under the curve (AUC) of the amide I band); and determining the value of the protein aggregation metric comprises calculating (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.

[0030] In certain embodiments, the one or more sample quality metrics further include a total protein content metric indicating the level of protein content within the aqueous sample.

[0031] In certain embodiments, step (d) comprises causing the processor to transmit one or more trigger signals (e.g., voltages) to a controller unit of the purification unit. In certain embodiments, the one or more trigger signals comprise analog voltage signals having a time-varying amplitude based on (e.g., substantially proportional to) the value of the protein aggregation metric. In certain embodiments, the one or more trigger signals comprise analog voltage signals having a time-varying amplitude based on (e.g., substantially proportional to) the value of the total protein content metric.

[0032] In certain embodiments, step (d) includes one or both of the following steps: initiating collection of the target fraction of the aqueous sample based on one or more trigger signals (e.g., 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 falls below a particular threshold); e.g., a particular one of the one or more trigger signals is a digital signal that triggers the initiation of collection of the target fraction (e.g., by transitioning from a voltage level of 0 to a voltage level of 1, or vice versa)); and stopping collection of the target fraction of the aqueous sample based on the one or more trigger signals (e.g., 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 falls below a particular threshold); e.g., a particular one of the one or more trigger signals is a digital signal that triggers the cessation of collection of the target fraction (e.g., by transitioning from a voltage level of 0 to a voltage level of 1, or vice versa)).

[0033] In another aspect, the disclosure provides a method for real-time monitoring of protein aggregation in a sample, comprising: (a) repeatedly receiving, by a processor of a computing device, infrared (IR) absorbance data corresponding to IR absorbance signals measured at each of a plurality of time points, the IR absorbance data including, for each particular time point of the plurality of time points, a corresponding IR absorbance spectrum measured from the sample at the particular time point, the IR absorbance data including a plurality of absorbance values, each associated with a particular wavenumber; (b) analyzing, by the processor (e.g., automatically), the IR absorbance data to, for each particular time point of the plurality of time points, determine a value of one or more peak metrics for one or both of an amide I band and an amide II band using the IR absorbance spectrum corresponding to the particular time point. determining a value of the one or more peak metrics; 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 a particular time point; and obtaining a real-time protein aggregation signal that provides a measure of protein aggregation in the sample as a function of time by updating the real-time protein aggregation signal with the determined value of the protein aggregation metric for the particular time point; and (c) storing and / or providing, by a processor, the real-time protein aggregation signal for one or more of: (i) further processing, (ii) display, and (iii) use as a control signal for regulating one or more purification units (e.g., a chromatography system).

[0034] In certain embodiments, step (b) comprises determining, for each particular time point, a value of a frequency location metric that quantifies the frequency (e.g., center of mass frequency, frequency of maximum height of the amide II band, center frequency of a fitted peak function (e.g., Gaussian, Lorentzian, etc.)) at which the amide II band is substantially centered at the particular time point as a protein aggregation metric value indicative of the level of protein aggregation in the sample at the particular time point. In certain embodiments, the frequency location metric is the center of mass frequency of the amide II band.

[0035] In certain embodiments, step (b) comprises, for each particular time point, 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, as the value of the protein aggregation metric, (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.

[0036] In another aspect, the present disclosure relates to a method for mid-IR (MIR) spectroscopy-based monitoring and control of a production unit for the manufacture of a biological product (e.g., a protein; e.g., a nucleic acid; e.g., a virus), comprising: (a) measuring, at each of one or more time points, by one or more (e.g., integrated) mid-infrared (MIR) analyzers (e.g., an MIR analyzer described in one or more aspects and / or embodiments (e.g., paragraphs above) herein), corresponding infrared (IR) absorbance signals from aqueous samples flowing to and / or from the production unit (e.g., and including one or more inputs, outputs, waste products, or ongoing products of the production unit); (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.

[0037] 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 flow or normal flow filtration unit, an ultrafiltration unit, and a diafiltration unit. In certain embodiments, the purification unit is a chromatography column {e.g., and 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 thereof (e.g., IEX and HIC; e.g., IEX and SEC)}.

[0038] In certain embodiments, the production unit is or includes a bioreactor (e.g., a seed bioreactor; e.g., a production bioreactor).

[0039] In certain embodiments, the aqueous sample comprises one or more protein species selected from the group consisting of a monoclonal antibody (mAb), a fusion protein, a viral capsid protein, an antibody-drug conjugate, a recombinant protein, and a plasma protein. In certain embodiments, the aqueous sample comprises peptide chains and / or protein fragments.

[0040] In certain embodiments, the aqueous sample contains a plurality of different protein species. In certain embodiments, the aqueous sample contains a heterogeneous population of target protein species, including monomeric and aggregated portions. In certain embodiments, the aqueous sample contains one or more subspecies of a particular protein species with a particular desired level and / or type of molecular conjugation (e.g., glycans, small molecule drugs, polyethylene glycol, etc.).

[0041] In certain embodiments, the aqueous sample comprises nucleic acids (eg, DNA, RNA, mRNA, etc.).

[0042] In certain embodiments, the aqueous sample comprises one or more species of virus and / or virus-like particle (e.g., adeno-associated viral vector (AAV); e.g., lentiviral vector).

[0043] In certain embodiments, step (c) includes using the IR absorbance data to determine values ​​of one or more sample quality metrics at each of the one or more time points (e.g., and adjusting one or more process parameters based thereon).

[0044] In certain embodiments, 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. In certain embodiments, the one or more sample quality metrics include a protein aggregation metric that indicates the level of protein aggregation in the aqueous sample. In certain embodiments, the one or more sample quality metrics include one or more protein species metrics that identify the presence and / or quantify the content (e.g., absolute content; e.g., relative content) of one or more specific protein species in the aqueous sample. In certain embodiments, the one or more sample quality metrics include a protein conjugation metric that quantifies the level and / or type of molecular conjugation (e.g., glycans, small molecule drugs, polyethylene glycol, etc.). In certain embodiments, 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., alpha helix content, beta sheet content, turn content, unfolded region content).

[0045] In certain embodiments, the one or more sample quality metrics include one or more nucleic acid content metrics that quantify the content of nucleic acids within the aqueous sample [e.g., 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., total and / or relative content of one or more specific types of nucleic acids (e.g., DNA, RNA, ssDNA, dsDNA)] (independent and / or distinguishable content metrics that measure viral nucleic acid and host nucleic acid, such as the total amount of a specific nucleotide base in a nucleic acid sample, such as GC content).

[0046] In certain embodiments, the one or more sample quality metrics include one or more virus content metrics (e.g., concentration (e.g., titer), mass per volume (e.g., mg / mL), number of (virus) particles per volume, etc.) that quantify the content of virus and / or virus-like particles within the aqueous sample. In certain embodiments, the one or more sample quality metrics include one or more empty / full capsid ratios (e.g., percent full viral vector, ratio, etc.) that quantify the content and / or relative proportion of empty and / or full viral vector within the aqueous sample. In certain embodiments, the one or more sample quality metrics include a capsid aggregation metric that indicates the level of capsid aggregation within the viral vector sample. In certain embodiments, the one or more sample quality metrics include a viral nucleic acid (e.g., viral DNA, RNA, etc.) content metric that distinguishes viral nucleic acid from host cell protein and host cell nucleic acid content.

[0047] 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 the IR spectral data {e.g., 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 quantifies a specific structural characteristic [e.g., intensity (e.g., peak amplitude; e.g., area under the curve (AUC)); e.g., linewidth; e.g., frequency location (e.g., peak frequency; e.g., center of mass frequency)] of one or more absorption peaks within the specific spectral bands}.

[0048] In certain embodiments, IR absorbance data is collected from (i) the amide II spectral band (e.g., at approximately 1500 cm -1 ~Approx. 1600cm -1 range; for example, about 1500 cm -1 ~Approx. 1575cm -1 range; for example, about 1500 cm -1 ~Approx. 1550cm -1 range; for example, about 1540 cm -1 ~Approx. 1560cm -1 and / or (ii) one or more (e.g., multiple) amide II absorbance values ​​that correlate with and measure IR absorption (of the virus sample) at wavenumbers within the range of about 1250 cm; and / or (iii) one or more (e.g., multiple) amide II absorbance values ​​that correlate with and measure IR absorption (of the virus sample) at wavenumbers within the range of about 1250 cm; and / or (iv) one or more (e.g., multiple) amide II absorbance values ​​that correlate with and measure IR absorption (of the virus sample) at wavenumbers within the range of about 1250 cm; and / or (v) one or more (e.g., multiple) amide II absorbance values ​​that correlate with and measure IR absorption (of the virus sample) at wavenumbers within the range of about 1250 cm; -1 ~Approx. 1350cm -1 range; for example, about 1250 cm -1 ~Approx. 1325cm -1 range; for example, about 1275 cm -1 ~Approx. 1325cm -1 range; for example, about 1280 cm -1 ~Approx. 1300cm -1 The amide III absorbance value includes one or more (e.g., multiple) amide III absorbance values ​​that relate to and measure IR absorption (of a virus sample) at wavenumbers within a range of 100 Hz to 100 Hz.

[0049] 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)) that quantifies protein content in the sample based at least in part on the amide II and / or amide III absorbance values. In certain embodiments, the method includes determining a value of an amide II peak metric based on the amide II absorbance values ​​and / or a value of an amide III peak metric based on the amide III absorbance values; and determining a protein content metric value using the amide II peak metric value and / or the amide III peak metric value. In certain embodiments, the amide II peak metric and / or the amide III peak metric are peak intensity metrics (e.g., peak height, area under the curve (AUC), etc.) that quantify the intensity of the amide II band and / or the amide III band, respectively.

[0050] In certain embodiments, IR absorbance data is collected from (i) the asymmetric PO spectral band (e.g., at approximately 1150 cm -1 ~Approx. 1250cm -1 range; for example, about 1175 cm -1 ~Approx. 1250cm -1 range; for example, about 1200 cm -1 ~Approx. 1250cm -1 range; for example, about 1210 cm -1 ~Approx. 1230cm -1 (ii) one or more (e.g., multiple) asymmetric phosphate stretch (asymmetric PO4) absorbance values ​​that correlate with and measure IR absorption (of the virus sample) at wavenumbers within a range of about 1000 cm-1; and / or (iii) one or more (e.g., multiple) asymmetric phosphate stretch (asymmetric PO4) absorbance values ​​that correlate with and measure IR absorption (of the virus sample) at wavenumbers within a range of about 1000 cm-1; and / or (iv) one or more (e.g., multiple) asymmetric phosphate stretch (asymmetric PO4) absorbance values ​​that correlate with and measure IR absorption (of the virus sample) at wavenumbers within a range of about 1000 cm-1; -1 ~Approx. 1100cm -1 range; for example, about 1050 cm -1 ~Approx. 1100cm -1 range; for example, about 1075 cm -1 ~Approx. 1100cm -1 range; for example, about 1075 cm -1 ~Approx. 1085cm -1The present invention includes one or more (e.g., multiple) symmetric phosphate extension (symmetric PO4) absorbance values ​​that are associated with and measure the IR absorption (of the virus sample) at wavenumbers within a range of 100 Hz to 100 Hz.

[0051] In certain embodiments, determining one or more sample quality metric values ​​includes determining a value of a nucleic acid (e.g., DNA, RNA, etc.) content metric (e.g., concentration (e.g., titer)) that quantifies nucleic acid content in the sample based at least in part on the asymmetric and / or symmetric P04 absorbance values. In certain embodiments, determining a value of an asymmetric P04 peak metric based on the asymmetric P04 absorbance values ​​and / or a value of a symmetric P04 peak metric based on the symmetric P04 absorbance values; and determining a nucleic acid content metric value using the asymmetric P04 peak metric value and / or the symmetric P04 peak metric value. In certain embodiments, the asymmetric P04 peak metric and / or the symmetric P04 peak metric are peak intensity metrics (e.g., peak height, area under the curve (AUC), etc.) that quantify the intensity of the asymmetric P04 band and / or the symmetric P04 band, respectively.

[0052] In certain embodiments, determining one or more sample quality metric values ​​involves independently quantifying total protein and nucleic acid content in the sample by determining (i) a value of a protein content metric [e.g., concentration (e.g., titer)] that quantifies protein content in the sample and (ii) a value of a nucleic acid (e.g., DNA, RNA, etc.) content metric that quantifies nucleic acid content [e.g., concentration (e.g., titer)] in the sample. In certain embodiments, determining one or more sample quality metric values ​​involves determining total capsid content based at least in part on (e.g., as a function of) the value of the protein content metric. In certain embodiments, determining one or more sample quality metric values ​​involves determining (i) the value of the protein content metric and / or total capsid content and (ii) the full capsid fraction based at least in part on (e.g., as a function of) the value of the nucleic acid content metric.

[0053] In certain embodiments, the IR absorbance data is or includes one or more IR absorbance spectra, each IR absorbance spectrum including, for each particular wavenumber of a plurality of wavenumbers across the measured spectral band, a corresponding IR absorbance value representing a measure of the absorption of IR light by the aqueous sample at the particular wavenumber.

[0054] In certain embodiments, the spectral bands measured span 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.

[0055] In certain embodiments, provided methods include monitoring one or more sample quality metrics over time by determining values ​​of the one or more sample quality metrics for each of one or more time points, hi certain embodiments, the methods include determining values ​​of the one or more sample quality metrics in substantially real time.

[0056] In certain embodiments, at least one specific sample quality metric of the one or more sample quality metrics is calculated using a machine learning model that receives one or more IR spectra as input and generates the specific sample quality metric as output. In certain embodiments, calculating the specific sample quality metric includes deconvolving the amide spectral region into sub-bands and / or calculating a second derivative spectrum.

[0057] In certain 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 measured from a corresponding (e.g., high-quality) reference sample (e.g., comprising a high purity and / or concentration of the target viral vector species and / or one or more model components thereof (e.g., a model protein solution; e.g., a model ssDNA solution)); and determining (e.g., automatically) values ​​for at least a portion of one or more sample quality metrics using the IR absorbance spectrum and the one or more reference spectra (e.g., determining one or more (e.g., multiple) measures of deviation between the reference spectrum and the IR absorbance spectrum as values ​​for a portion of one or more viral vector sample quality metrics).

[0058] In certain embodiments, the one or more reference spectra comprise high-quality viral vector spectra measured from a reference sample having a full capsid fraction at or above a certain threshold fraction (e.g., known a priori; e.g., determined). In certain embodiments, the threshold fraction is about 75% [e.g., about 80% (e.g., about 90%)].

[0059] 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 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). In certain embodiments, step (c) includes calculating one or more derivative spectra (e.g., first derivative; e.g., second derivative) of at least one of the one or more reference spectra and / or the IR absorbance spectrum. In certain embodiments, step (c) comprises calculating (e.g., as values ​​for one or more sample quality metrics) one or more members selected from the group consisting of: (i) a correlation value based on the correlation between one or more reference spectra and / or a particular one of their one or more derivatives and (ii) the IR absorbance spectrum and / or one or more derivatives thereof; a covariance value based on the covariance between (i) the one or more reference spectra and / or a particular one of their one or more derivatives and (ii) the IR absorbance spectrum and / or one or more derivatives thereof; a Pearson correlation value between (i) the one or more reference spectra and / or a particular one of their one or more derivatives and (ii) the IR absorbance spectrum and / or one or more derivatives thereof; and an overlap integral value based on the overlap integral between (i) the one or more reference spectra and / or a particular one of their one or more derivatives and (ii) the IR absorbance spectrum and / or one or more derivatives thereof.

[0060] In certain embodiments, step (c) includes obtaining a set of sample peak metric values ​​by determining values ​​for a set of one or more specific peak metrics from the IR absorbance spectrum; and determining one or more values ​​of sample quality metrics based on the set of sample peak metric values ​​and a set of reference peak metric values ​​determined for the one or more specific peak metrics from one or more reference spectra.

[0061] In certain embodiments, provided methods include determining a similarity score (e.g., as one or more sample quality metrics) measuring the similarity between one or more reference spectra and the IR absorbance spectrum. In certain embodiments, the method includes monitoring deviations from the one or more reference spectra in real time by repeatedly performing steps (a)-(c) in substantially real time.

[0062] In certain embodiments, the one or more time points are multiple time points (e.g., step (a) comprises (e.g., repeatedly) measuring the IR absorption signal at each of multiple time points (e.g., continuously, in real time)).

[0063] In certain embodiments, a provided method includes determining a value of a first sample quality metric at each of a plurality of time points, and determining a value of a second (e.g., time-differential; e.g., time-aggregated) sample quality metric using values ​​of the first sample quality metric corresponding to two or more of the plurality of time points.

[0064] In certain embodiments, the second sample quality metric is a time difference metric that measures the temporal change of the first sample quality metric and is calculated based on the difference between (i) the value of the first sample quality metric at a first set of time points and (ii) the value of the first sample quality metric at 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 values ​​at consecutive time points)).

[0065] In certain embodiments, the second sample quality metric is a (e.g., real-time) time-aggregated signal that is a function (e.g., cumulative sum, mean, median, mode, variance, standard deviation, etc. over a particular time window) of at least a portion of multiple time points (e.g., a cumulative, increasing portion 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 backward window)).

[0066] In certain embodiments, step (c) includes causing the processor to transmit one or more trigger signals (e.g., voltages) to a controller unit of the production unit. In certain embodiments, the one or more trigger signals include analog voltage signals having time-varying amplitudes based on (e.g., substantially proportional to) at least in part the values ​​of the one or more sample quality metrics.

[0067] In certain embodiments, step (c) comprises using a machine learning model to adjust one or more process parameters (e.g., the machine learning model receives one or more sample quality metrics as inputs and generates adjustments to and / or target process parameters as outputs; e.g., the machine learning model receives one or more IR spectra as inputs and generates adjustments to and / or target process parameters as outputs).

[0068] In certain 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 certain embodiments, the one or more process parameters include the amount (e.g., absolute and / or relative) of one or more raw materials (e.g., used as input to a production unit). In certain embodiments, the one or more process parameters include the time to start and / or stop a subprocess (e.g., heating, collection of elution fractions, growth, etc.).

[0069] In certain embodiments, one or more IR absorbance signals, to which the IR absorbance data received in step (b) correspond, are measured from the aqueous sample at each of one or more time points as the aqueous sample exits a purification unit (e.g., a chromatography column); the method includes using the IR absorbance data to determine one or both of (i) total capsid content and (ii) full capsid fraction; and using the determined total capsid content and / or full capsid fraction to control collection of target fractions of the aqueous sample (e.g., during a particular collection window) thereby obtaining a purified sample of the viral vector material.

[0070] In certain embodiments, one or more IR absorbance signals, to which the IR absorbance data received in step (b) correspond, are measured from the aqueous sample at each of one or more time points as the aqueous sample exits a purification unit (e.g., a chromatography column); the method includes using the IR absorbance data to determine a capsid aggregation metric that measures the level of aggregation between capsids in the viral vector sample; and obtaining a purified sample of viral vector material by using the capsid aggregation metric to control collection of a target fraction of the aqueous sample (e.g., during a specific collection window).

[0071] In another aspect, the disclosure provides a system for obtaining a purified sample of a target protein species with real-time monitoring of protein heterogeneity and (e.g., automated; e.g., semi-automated) control of the purification process, the system comprising: (a) one or more mid-infrared (MIR) analyzers aligned and operable to measure, at each of one or more time points, a corresponding infrared (IR) absorbance signal from 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) a processor of a computing device; and (c) a memory having instructions stored thereon; The instructions, when executed by a processor, relate to a system that causes the processor to receive IR absorbance data corresponding to the IR absorbance signal at each of one or more time points; determine values ​​of one or more sample quality metrics based on the IR absorbance spectrum, including one or more sample quality metrics, including a protein aggregation metric indicative of a level of protein aggregation within the aqueous sample; and provide (e.g., send to a controller unit of the purification unit) and / or use the one or more sample quality metrics to control collection of a target fraction of the aqueous sample (e.g., during a particular collection window), thereby obtaining a purified sample of the target protein species.

[0072] In certain embodiments, the provided system further comprises a purification unit and / or its controller unit.

[0073] In another aspect, the disclosure provides a system for real-time monitoring of protein aggregation in a sample, the system including: a processor of a computing device; and a memory having stored thereon instructions, which, when executed by the processor, cause the processor to (a) repeatedly receive infrared (IR) absorbance data corresponding to an IR absorbance signal measured at each of a plurality of time points, the IR absorbance data including, for each particular time point of the plurality of time points, a corresponding IR absorbance spectrum measured from the sample at the particular time point, the IR absorbance data including a plurality of absorbance values ​​each associated with a particular wavenumber; and (b) analyze (e.g., automatically) the IR absorbance data to, for each particular time point of the plurality of time points, identify one of an amide I band and an amide II band using the IR absorbance spectrum corresponding to the particular time point. or both; 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 a particular time point; and updating the real-time protein aggregation signal with the determined value of the protein aggregation metric for the particular time point, thereby obtaining a real-time protein aggregation signal that provides a measure of protein aggregation in the sample as a function of time; (c) storing and / or providing the real-time protein aggregation signal for one or more of: (i) further processing, (ii) display, and (iii) use as a control signal for adjusting one or more purification units (e.g., a chromatography system).

[0074] In another aspect, the present disclosure relates to a system for mid-IR (MIR) spectroscopy-based monitoring and control of a production unit for the manufacture of a biological product (e.g., a protein; e.g., a virus), comprising: (a) one or more (e.g., integrated) mid-infrared (MIR) analyzers (e.g., an MIR analyzer described in one or more aspects and / or embodiments (e.g., the paragraph above) of the present specification) aligned and operable to measure, at each of one or more time points, a corresponding IR absorbance signal from an aqueous sample flowing to and / or from the production unit (e.g., and including one or more inputs, outputs, waste products, or ongoing products of the production unit); (b) a processor of a computing device; and (c) a memory having instructions stored thereon that, 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 adjust one or more process parameters of the production unit.

[0075] In certain embodiments, the provided system further comprises a production unit and / or its controller unit.

[0076] In another aspect, the present disclosure relates to a method for quantifying and / or monitoring (e.g., in real time) viral vector quality in an aqueous sample containing one or more species of virus and / or virus-like particles, the method comprising: (a) receiving (e.g., repeatedly), by a processor of a computing device, infrared (IR) absorbance data corresponding to one or more IR absorbance signals measured from the sample; (b) determining (e.g., automatically), by the processor, values ​​for one or more viral vector sample quality metrics using the IR absorbance data; and (c) storing and / or providing the one or more viral vector sample quality metric values ​​for display and / or further processing.

[0077] In certain embodiments, the one or more viral vector quality metrics include a total capsid content (e.g., concentration (e.g., titer), mass per volume (e.g., mg / mL), number of (viral) particles per volume, etc.) that quantifies the content of viral capsids in a sample. In certain embodiments, the one or more viral vector sample quality metrics include a full capsid fraction (e.g., percent full viral vector, ratio, etc.). In certain embodiments, the one or more viral vector sample quality metrics include a capsid aggregation metric that indicates the level of capsid aggregation in a viral vector sample. In certain embodiments, the one or more viral vector sample quality metric includes a protein content metric (e.g., concentration (e.g., titer), mass per volume (e.g., mg / mL), number of particles per volume, etc.) that quantifies the protein content in a sample.

[0078] In certain embodiments, the one or more viral vector sample quality metrics include 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), mass per volume (e.g., mg / mL), number of particles per volume, viral genome copy number, etc.).

[0079] In certain embodiments, the one or more viral vector sample quality metrics include a viral nucleic acid (e.g., viral DNA, RNA, etc.) content metric that distinguishes viral nucleic acid from host cell protein and host cell nucleic acid content.

[0080] In certain embodiments, step (b) comprises determining, by a processor, values ​​for each of one or more peak metrics of the IR absorption data, each peak metric associated with one or more particular 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 amide bands; e.g., an asymmetric PO4 band; e.g., a symmetric PO4 band)), and quantifying particular structural characteristics (e.g., intensity (e.g., peak amplitude; e.g., area under the curve (AUC)); e.g., linewidth; e.g., frequency location (e.g., peak frequency; e.g., center of mass frequency)) of one or more absorption peaks within the particular spectral bands; and using the determined values ​​of the one or more peak metrics to determine values ​​of at least a portion of the viral vector sample quality metrics.

[0081] In certain embodiments, IR absorbance data is collected from (i) the amide II spectral band (e.g., at approximately 1500 cm -1 ~Approx. 1600cm -1 range; for example, about 1500 cm -1 ~Approx. 1575cm -1 range; for example, about 1500 cm -1 ~Approx. 1550cm -1 range; for example, about 1540 cm -1 ~Approx. 1560cm -1 and / or (ii) one or more (e.g., multiple) amide II absorbance values ​​that correlate with and measure IR absorption (of the virus sample) at wavenumbers within the range of about 1250 cm; and / or (iii) one or more (e.g., multiple) amide II absorbance values ​​that correlate with and measure IR absorption (of the virus sample) at wavenumbers within the range of about 1250 cm; and / or (iv) one or more (e.g., multiple) amide II absorbance values ​​that correlate with and measure IR absorption (of the virus sample) at wavenumbers within the range of about 1250 cm; and / or (v) one or more (e.g., multiple) amide II absorbance values ​​that correlate with and measure IR absorption (of the virus sample) at wavenumbers within the range of about 1250 cm; -1 ~Approx. 1350cm -1 range; for example, about 1250 cm -1 ~Approx. 1325cm -1 range; for example, about 1275 cm -1 ~Approx. 1325cm -1 range; for example, about 1280 cm -1 ~Approx. 1300cm -1The amide III absorbance value includes one or more (e.g., multiple) amide III absorbance values ​​that relate to and measure IR absorption (of a virus sample) at wavenumbers within a range of 100 Hz to 100 Hz.

[0082] In certain embodiments, step (b) includes determining a value of a protein content metric [e.g., concentration (e.g., titer)] that quantifies the protein content in the sample based at least in part on the amide II and / or amide III absorbance values.

[0083] In certain embodiments, provided methods include determining an amide II peak metric value based on the amide II absorbance value and / or an amide III peak metric value based on the amide III absorbance value; and determining a protein content metric value using the amide II peak metric value and / or the amide III peak metric value. In certain embodiments, the amide II peak metric and / or the amide III peak metric are peak intensity metrics (e.g., peak height, area under the curve (AUC), etc.) that quantify the intensity of the amide II band and / or the amide III band, respectively.

[0084] In certain embodiments, 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 abundance (e.g., absolute abundance; e.g., relative abundance) of one or more particular protein structural forms (e.g., particular secondary structural motifs; e.g., particular tertiary and / or quaternary structural motifs / forms) within the sample (e.g., thereby providing monitoring of variations in capsid protein secondary / tertiary / quaternary structure).

[0085] In certain embodiments, IR absorbance data is collected from (i) the asymmetric PO spectral band (e.g., at approximately 1150 cm -1 ~Approx. 1250cm -1 range; for example, about 1175 cm -1 ~Approx. 1250cm-1 range; for example, about 1200 cm -1 ~Approx. 1250cm -1 range; for example, about 1210 cm -1 ~Approx. 1230cm -1 (ii) one or more (e.g., multiple) asymmetric phosphate stretch (asymmetric PO4) absorbance values ​​that correlate with and measure IR absorption (of the virus sample) at wavenumbers within a range of about 1000 cm-1; and / or (iii) one or more (e.g., multiple) asymmetric phosphate stretch (asymmetric PO4) absorbance values ​​that correlate with and measure IR absorption (of the virus sample) at wavenumbers within a range of about 1000 cm-1; and / or (iv) one or more (e.g., multiple) asymmetric phosphate stretch (asymmetric PO4) absorbance values ​​that correlate with and measure IR absorption (of the virus sample) at wavenumbers within a range of about 1000 cm-1; -1 ~Approx. 1100cm -1 range; for example, about 1050 cm -1 ~Approx. 1100cm -1 range; for example, about 1075 cm -1 ~Approx. 1100cm -1 range; for example, about 1075 cm -1 ~Approx. 1085cm -1 The present invention includes one or more (e.g., multiple) symmetric phosphate extension (symmetric PO4) absorbance values ​​that are associated with and measure the IR absorption (of the virus sample) at wavenumbers within a range of 100 Hz to 100 Hz.

[0086] In certain embodiments, step (b) comprises determining a value of a nucleic acid (e.g., DNA, RNA, etc.) content metric (e.g., concentration (e.g., titer)) that quantifies nucleic acid content in the sample based at least in part on the asymmetric and / or symmetric P04 absorbance values. In certain embodiments, provided methods comprise determining a value of an asymmetric P04 peak metric based on the asymmetric P04 absorbance values ​​and / or a value of a symmetric P04 peak metric based on the symmetric P04 absorbance values; and determining a nucleic acid content metric value using the asymmetric P04 peak metric value and / or the symmetric P04 peak metric value.

[0087] In certain embodiments, the asymmetric PO4 peak metric and / or the symmetric PO4 peak metric are peak intensity metrics (e.g., peak height, area under the curve (AUC), etc.) that quantify the intensity of the asymmetric PO4 band and / or the symmetric PO4 band, respectively.

[0088] In certain embodiments, step (b) comprises independently quantifying total protein and nucleic acid content in the sample by determining (i) a value of a protein content metric [e.g., concentration (e.g., titer)] that quantifies protein content in the sample and (ii) a value of a nucleic acid (e.g., DNA, RNA, etc.) content metric that quantifies nucleic acid content [e.g., concentration (e.g., titer)] in the sample. In certain embodiments, step (b) comprises determining total capsid content as one of the viral vector sample quality metrics based at least in part on (e.g., as a function of) the value of the protein content metric. In certain embodiments, step (b) comprises determining full capsid fraction as one of the viral vector sample quality metrics based at least in part on (e.g., as a function of) the value of (i) the protein content metric and / or total capsid content and (ii) the value of the nucleic acid content metric.

[0089] In certain embodiments, the IR absorbance data is or includes one or more IR absorbance spectra, each IR absorbance spectrum including, for each particular wavenumber of a plurality of wavenumbers across a measured spectral band, a corresponding IR absorbance value that represents a measure of the absorption of IR light by the aqueous sample at the particular wavenumber. In certain embodiments, the measured spectral bands span one or more bands selected from the group consisting of an amide II band, an amide III band, an asymmetric PO4 band, and a symmetric PO4 band.

[0090] In certain embodiments, step (a) includes repeatedly receiving IR absorbance data at multiple time points to obtain a corresponding set of IR absorbance data for each of the multiple time points, and the provided method includes monitoring the total capsid content and / or full capsid fraction over time by performing steps (b)-(c) for each set of IR absorbance data.

[0091] In certain embodiments, provided methods include performing steps (a)-(c) substantially in real time to obtain (i) a real-time capsid content signal that provides a measure of capsid content in the sample as a function of time and / or a full capsid fraction signal that provides a measure of the fraction of capsids in the sample that are full as a function of time.

[0092] In certain embodiments, the provided methods include measuring one or more IR absorbance signals by one or more (e.g., integrated) mid-infrared (MIR) analyzers (e.g., the MIR analyzers described in various aspects and / or embodiments herein (e.g., the paragraphs above)).

[0093] In certain embodiments, provided methods include measuring, at each of one or more time points, a corresponding one of one or more infrared (IR) absorbance signals.

[0094] In certain embodiments, provided methods include measuring one or more IR absorbance signals derived from an aqueous sample as the aqueous sample flows to (e.g., into) and / or from a production unit (e.g., the aqueous sample comprises one or more inputs, outputs, waste products, or ongoing products of a production unit).

[0095] In certain embodiments, the one or more species of virus and / or virus-like particle comprises one or more species of adeno-associated virus (AAV). In certain embodiments, the one or more species of virus comprises adenovirus and / or retrovirus (e.g., lentivirus). In certain embodiments, the one or more species of virus comprises a plant-based virus (e.g., tobacco mosaic virus).

[0096] In certain embodiments, one or more IR absorbance signals to which the IR absorbance data corresponds are measured from the aqueous sample as it flows to (e.g., into) and / or from a production unit (e.g., the aqueous sample comprises one or more inputs, outputs, waste products, or ongoing products of the production unit).

[0097] In certain embodiments, the production unit is 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 flow or normal flow filtration unit, an ultrafiltration unit, and a diafiltration unit. In certain embodiments, the purification unit is a chromatography column {e.g., and 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 thereof (e.g., IEX and HIC; e.g., IEX and SEC)}.

[0098] In certain embodiments, the production unit is or comprises a bioreactor (e.g., a seed bioreactor; e.g., a production bioreactor).

[0099] In certain embodiments, step (c) comprises causing, by the processor, generation and / or transmission of one or more trigger signals (e.g., voltages) to a control unit of the production unit based at least in part on one or more determined viral vector sample quality metrics (e.g., values ​​thereof) (e.g., determined capsid content value and / or determined full capsid percentage (e.g., values ​​thereof)). In certain embodiments, step (c) comprises causing, by the processor, generation of a trigger signal (e.g., analog signal) having a value based at least in part on the determined viral vector quality metrics (e.g., capsid content; e.g., full capsid percentage; e.g., capsid aggregation metrics).

[0100] In certain embodiments, one or more IR absorbance signals to which the IR absorbance data received in step (a) correspond are measured from the aqueous sample at each of one or more time points as the aqueous sample exits a purification unit (e.g., a chromatography column); the provided method includes using the IR absorbance data to determine one or both of (i) total capsid content and (ii) full capsid fraction; and using the determined total capsid content and / or full capsid fraction to control collection of target fractions of the aqueous sample (e.g., during a specific collection window) thereby obtaining a purified sample of viral vector material.

[0101] In certain embodiments, one or more IR absorbance signals, to which the IR absorbance data received in step (a) correspond, are measured from the aqueous sample at each of one or more time points as the aqueous sample exits a purification unit (e.g., a chromatography column); the provided method includes using the IR absorbance 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 collection of a 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, the IR absorbance data includes an IR absorbance spectrum, and step (b) includes receiving (e.g., and / or accessing) one or more reference spectra, each measured from a corresponding (e.g., high-quality) reference sample comprising the target viral vector species and / or one or more model components thereof (e.g., a model protein solution; e.g., a model ssDNA solution) of high purity and / or concentration; and determining (e.g., automatically) values ​​for at least a portion of one or more viral vector sample quality metrics using the IR absorbance spectrum and the one or more reference spectra (e.g., determining one or more (e.g., multiple) measures of deviation between the reference spectrum and the IR absorbance spectrum as values ​​for a portion of the one or more viral vector sample quality metrics).

[0103] In certain embodiments, the one or more reference spectra comprise high-quality viral vector spectra measured from a reference sample having a full capsid fraction at or above a certain threshold fraction (e.g., known a priori; e.g., determined). In certain embodiments, the threshold fraction is about 75% [e.g., about 80% (e.g., about 90%)].

[0104] 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 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).

[0105] In certain embodiments, step (b) includes calculating one or more derivative spectra (e.g., first derivative; e.g., second derivative) of at least one of the one or more reference spectra and / or the IR absorbance spectrum.

[0106] In certain embodiments, step (b) comprises calculating (e.g., as a measure of deviation) one or more members selected from the group consisting of: (i) a correlation value based on the correlation between one or more reference spectra and / or a particular one of their one or more derivatives and (ii) the IR absorbance spectrum and / or one or more derivatives thereof; a covariance value based on the covariance between (i) the one or more reference spectra and / or a particular one of their one or more derivatives and (ii) the IR absorbance spectrum and / or one or more derivatives thereof; a Pearson correlation value between (i) the one or more reference spectra and / or a particular one of their one or more derivatives and (ii) the IR absorbance spectrum and / or one or more derivatives thereof; and an overlap integral value based on the overlap integral between (i) the one or more reference spectra and / or a particular one of their one or more derivatives and (ii) the IR absorbance spectrum and / or one or more derivatives thereof.

[0107] In certain embodiments, step (b) includes obtaining a set of sample peak metric values ​​by determining values ​​for a set of one or more specific peak metrics from the IR absorbance spectrum; and determining a measure of deviation based on the set of sample peak metric values ​​and a set of reference peak metric values ​​determined for the one or more specific peak metrics from one or more reference spectra.

[0108] In certain embodiments, the provided method includes determining a similarity score that measures the similarity between one or more reference spectra and the IR absorbance spectrum as a measure of deviation.

[0109] In certain embodiments, the provided method includes monitoring deviations from one or more reference spectra in real time by repeatedly performing steps (a)-(c) in substantially real time.

[0110] In another aspect, the disclosure provides a method for assessing and / or monitoring (e.g., in real time) the quality of viral vector content within an aqueous sample containing a target viral vector species, the method comprising: (a) receiving (e.g., repeatedly), by a processor of a computing device, infrared (IR) absorbance data corresponding to one or more IR absorbance signals measured from the sample, the IR absorbance data comprising (e.g., at least one) IR absorbance spectrum measured from the sample, the IR absorbance data including a plurality of absorbance values ​​each associated with a particular wavenumber; (b) determining, by the processor, whether or not the target viral vector is of high purity and / or concentration; (c) receiving (e.g., and / or accessing) one or more reference spectra, each measured from a corresponding (e.g., high-quality) reference sample comprising the IR absorbance spectrum of the target molecule and / or one or more model components thereof (e.g., a model protein solution; e.g., a model ssDNA solution); (d) determining (e.g., automatically), by a processor, one or more (e.g., multiple) deviation measures using the IR absorbance spectrum and the one or more reference spectra; and (d) storing and / or providing the deviation measures for display and / or further processing.

[0111] In another aspect, the present disclosure relates to a system for quantifying and / or monitoring (e.g., in real time) viral vector quality in an aqueous sample containing one or more species of virus and / or virus-like particles, the system including: a processor of a computing device; and a memory having stored thereon instructions that, when executed by the processor, cause the processor to perform provided methods described in certain aspects and embodiments herein (e.g., in the paragraphs above).

[0112] In another aspect, the disclosure provides a method for (e.g., real-time) monitoring of compositional changes in a sample by infrared (IR) absorption spectroscopy, comprising: (a) repeatedly receiving, by a processor of a computing device, infrared (IR) absorbance data corresponding to IR absorbance signals measured at each of a plurality of time points, the IR absorbance data including, for each particular time point of the plurality of time points, a corresponding IR absorbance spectrum measured from the sample at the particular time point, the IR absorbance data including a plurality of absorbance values, each associated with a particular wavenumber; (b) analyzing, by the processor (e.g., automatically) the IR absorbance data to obtain, for each particular time point of the plurality of time points, a current normalized spectrum by normalizing the current IR absorbance spectrum corresponding to the particular time point using reference absorbance values ​​determined from values ​​of the current IR absorbance spectrum at one or more reference wavenumbers; and determining a current value of a spectral difference metric based on the difference between the current normalized spectrum and a prior normalized spectrum. wherein the previous normalized spectrum corresponds to and is based on one or more previously obtained IR absorption spectra (e.g., a specific previously obtained IR absorption spectrum; e.g., an average of multiple previously obtained IR absorption spectra) measured at a specific previous time point (e.g., a specific time interval before the current specific time point and / or multiple thereof), and each specific previously obtained IR absorption spectrum has been normalized using a reference value determined from values ​​of the specific previously obtained IR absorbance spectrum at one or more reference wavenumbers; and obtaining a (e.g., real-time) normalized spectral difference signal measuring a change in normalized spectral absorbance between successive time points by updating the real-time normalized spectral difference signal according to a current value of the normalized spectral difference metric; and (c) transmitting, by a processor, (i) further processing, (ii) displaying, and (iii) transmitting the normalized spectral difference signal to a production unit (e.g., a chromatography unit;storing and / or providing a real-time normalized spectral difference signal for one or more of: use as a control signal for adjustment of one or more process parameters of a process (e.g., a filtration unit);

[0113] In certain embodiments, the provided methods further include identifying (e.g., by a processor) a change in the composition of the sample (e.g., at a particular time) based on the real-time normalized spectral difference signal.

[0114] In certain embodiments, provided methods include detecting change points (e.g., statistical characteristics of the change (e.g., mean, median, mode, variance, etc.); e.g., step changes) in a real-time normalized spectral difference signal and identifying a change in the composition of the sample based on the detected change points.

[0115] In certain embodiments, the provided methods include determining the value of one or more statistical parameters (e.g., as a sample quality metric) of the real-time normalized spectral difference signal (e.g., at each time point, e.g., in real time).

[0116] In certain embodiments, the one or more statistical parameters comprise one or more members selected from the group consisting of: mean (e.g., calculated as the average of the real-time normalized spectral difference signal over a time window that includes (e.g., ends at) the current time point and one or more previous time points, e.g., a running (e.g., backward) mean); variance (e.g., calculated as the variance of the real-time normalized spectral difference signal over a time window that includes (e.g., ends at) the current time point and one or more previous time points, e.g., a running (e.g., backward) variance); mode; and standard deviation.

[0117] In certain embodiments, the provided methods include identifying a change in composition based on the value of at least one of the one or more statistical parameters (i) exceeding one or more thresholds and / or (ii) changing outside a particular range (e.g., a predetermined threshold and / or range; e.g., a threshold and / or range determined ad hoc during an early stage of a process run (e.g., during an early time window of a chromatography run, e.g., during an initial increase in the salt gradient before the protein elutes)).

[0118] In certain embodiments, the sample is or comprises an aqueous sample, hi certain embodiments, the sample comprises one or more protein species (e.g., a target protein species such as, e.g., a monoclonal antibody, as described in certain embodiments herein).

[0119] In certain embodiments, the provided methods further include identifying (e.g., by a processor) a change in the composition of the sample (e.g., at a particular time) that corresponds to a change in the purity (e.g., the presence of non-target protein species; e.g., the presence of target protein species in an undesirable form (e.g., non-monomeric)) and / or properties (e.g., secondary structure composition) of the target protein species.

[0120] In certain embodiments, the identified change in composition is, or includes (e.g., indicates) a change in the level and / or presence of protein aggregates within the aqueous sample. In certain embodiments, the identified change in composition 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., glycans, small molecule drugs, polyethylene glycol, etc.); and a change in the content of one or more protein secondary structure motifs (e.g., alpha helix content, beta sheet content, turn content, denatured region content).

[0121] In certain embodiments, the sample comprises nucleic acids (e.g., DNA, RNA, mRNA, etc.). In certain embodiments, provided methods further include identifying (e.g., by a processor) a change in the composition of the sample (e.g., at a particular time) corresponding to a change in the purity and / or characteristics of nucleic acids within the sample (e.g., a change in the relative content of one or more particular types of nucleic acids (e.g., DNA, RNA, ssDNA, dsDNA) within the (e.g., aqueous) sample).

[0122] In certain embodiments, the aqueous sample contains one or more species of virus and / or virus-like particles (e.g., adeno-associated viral vectors (AAV); e.g., lentiviral vectors). In certain embodiments, the provided methods further include identifying (e.g., by a processor) a change in the composition of the sample (e.g., at a particular time) that corresponds to a change in the purity and / or characteristics of the virus and / or virus-like particles within the sample.

[0123] In certain embodiments, the change in composition corresponds to (e.g., indicates) a change in the relative proportion (e.g., percent, ratio, etc., of full viral vector) of empty and / or full viral vector within the aqueous sample.

[0124] In certain embodiments, the change in composition corresponds to (e.g., is indicative of) the level of capsid aggregation within the sample.

[0125] In certain embodiments, the change in composition corresponds to (e.g., indicates) a change in the relative content between viral nucleic acid derived from host cell proteins and host cell nucleic acid content.

[0126] In certain embodiments, the provided methods include causing (e.g., inducing) adjustment of one or more process parameters of a production unit based on (e.g., triggered upon) identification of a change in the composition of 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).

[0127] In certain embodiments, the production unit is or includes a purification unit.

[0128] In certain embodiments, the purification unit is or comprises a chromatography column.

[0129] 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., temporal) profiles).

[0130] In certain embodiments, the provided methods include eliciting a response (e.g., any of the parameters discussed herein) of a production unit (e.g., the production unit is a first production unit and the sample is combined with (e.g., is an input, output, or component processed by) a second (e.g., upstream or downstream) production unit).

[0131] In certain embodiments, the sample is an aqueous sample, and the method includes obtaining a purified sample by causing adjustment of a collection window to control collection of a target fraction (e.g., a mono- or mono-atomic species of protein) of the aqueous sample.

[0132] In certain embodiments, the production unit is or includes a filtration unit (e.g., an ultrafiltration and / or diafiltration unit) (e.g., and the method includes steps that cause adjustments to flow rate, transmembrane pressure, processing time, etc. to control the composition of the retentate and / or filtration product).

[0133] In certain embodiments, the method includes monitoring the progress of a chemical reaction (e.g., within a production unit (e.g., a bioreactor, a transfection unit, a PEGylation unit, an antibody-drug conjugation unit)) based on a real-time normalized spectral difference signal (e.g., identifying compositional changes using the real-time normalized spectral difference signal).

[0134] In certain embodiments, the method includes a step of causing adjustment of one or more members selected from the group consisting of in-line buffer preparation, a mixing process (e.g., in a mixing tank), a temperature controller.

[0135] In certain embodiments, the reference absorbance value is determined from values ​​of a current IR absorbance spectrum at a single reference wavenumber, and the previous reference value is determined from values ​​of a previous IR absorbance spectrum at a single reference wavenumber.

[0136] In certain embodiments, calculating the current spectral difference metric includes calculating (e.g., subtracting and subtracting) the integrated absorbance values ​​across one or more specific spectral bands for each of the current normalized spectrum and the previous normalized spectrum.

[0137] In certain embodiments, the one or more specific spectral bands are the amide I spectral band (e.g., at about 1600 cm -1 ~Approx. 1700cm -1 or approximately 1800 cm -1 range (e.g., about 1600 cm -1 ~Approx. 1725cm-1 range; for example, about 1625 cm -1 ~Approx. 1725cm -1 range; for example, about 1630 cm -1 ~Approx. 1650cm -1 range), and / or about 1500 to about 1600 cm -1 range (e.g., about 1500 cm -1 ~Approx. 1575cm -1 range; for example, about 1500 cm -1 ~Approx. 1550cm -1 range; for example, about 1540 cm -1 ~Approx. 1560cm -1 amide II region; amide II spectral bands (e.g., around 1500 cm -1 ~Approx. 1600cm -1 range; for example, about 1500 cm -1 ~Approx. 1575cm -1 range; for example, about 1500 cm -1 ~Approx. 1550cm -1 range; for example, about 1540 cm -1 ~Approx. 1560cm -1 range); and amide III spectral bands (e.g., approximately 1250 cm -1 ~Approx. 1350cm -1 range; for example, about 1250 cm -1 ~Approx. 1325cm -1 range; for example, about 1275 cm -1 ~Approx. 1325cm -1 range; for example, about 1280 cm -1 ~Approx. 1300cm -1 The range of the number of members selected from the group consisting of:

[0138] In certain embodiments, the one or more particular spectral bands are asymmetric PO spectral bands (e.g., at about 1150 cm -1 ~Approx. 1250cm -1 range; for example, about 1175 cm -1 ~Approx. 1250cm -1 range; for example, about 1200 cm -1 ~Approx. 1250cm -1range; for example, about 1210 cm -1 ~Approx. 1230cm -1 range); and symmetric PO4 spectral bands (e.g., around 1000 cm -1 ~Approx. 1100cm -1 range; for example, about 1050 cm -1 ~Approx. 1100cm -1 range; for example, about 1075 cm -1 ~Approx. 1100cm -1 range; for example, about 1075 cm -1 ~Approx. 1085cm -1 The range of the number of members selected from the group consisting of:

[0139] In certain embodiments, the method further comprises measuring the IR absorbance signal using one or more MIR analyzers (e.g., as described in any one of claims 9 to 24).

[0140] In another aspect, the disclosure provides a method for monitoring temporal changes (e.g., real-time) of a sample by infrared (IR) absorption spectroscopy, comprising: (a) repeatedly receiving, by a processor of a computing device, infrared (IR) absorbance data corresponding to IR absorbance signals measured at each of a plurality of time points, wherein the IR absorbance data includes, for each particular time point of the plurality of time points, a corresponding IR absorbance spectrum measured from the sample at the particular time point, the IR absorbance data including a plurality of absorbance values, each associated with a particular wavenumber; (b) analyzing (e.g., automatically) the IR absorbance data by the processor to determine one or more properties (e.g., sample quality) determined using (i) a first set of one or more IR absorbance spectra corresponding to the 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 (c) storing and / or providing, by a processor, the time difference signal and / or the time aggregated signal for one or more of: (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).

[0141] In another aspect, the present disclosure provides a system for (e.g., real-time) monitoring of temporal (e.g., compositional) changes in a sample by infrared (IR) absorption spectroscopy, the system including: a processor of a computing device; and a memory having stored thereon instructions that, when executed by the processor, cause the processor to perform various methods described herein (e.g., in the paragraph above).

[0142] In certain embodiments, the provided system further includes one or more MIR analyzers (eg, as described in the paragraph above).

[0143] In some aspects, the present disclosure provides methods for obtaining a purified sample of a target protein species through mid-infrared (IR) spectroscopy-based bioprocess monitoring and control, the provided method including: (a) measuring, by one or more mid-infrared (IR) analytical devices, a plurality of mid-IR absorbance spectra over time by measuring, at each of a plurality of time points, a corresponding mid-IR absorbance spectrum from an aqueous sample exiting a purification unit, the aqueous sample comprising one or more protein species, including the target protein species; (b) receiving, by a processor of a computing device, spectral data corresponding to the plurality of measured mid-IR absorbance spectra; (c) determining, by the processor, for at least some of the plurality of time points based on the spectral data, corresponding values ​​of one or more sample quality metrics comprising measures of the concentration and / or purity of the target protein species in the aqueous sample; and (d) obtaining a purified sample of the target protein species by using the determined values ​​of the one or more sample quality metrics to control the collection of target fractions of the aqueous sample (e.g., during a particular collection window).

[0144] In certain embodiments, the purification unit is or comprises a chromatography column {e.g., 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 thereof (e.g., IEX and HIC; e.g., IEX and SEC)}. In certain embodiments, the purification unit is or comprises an ultrafiltration and diafiltration system (UF / DF) (e.g., a tangential flow filtration (TFF) system).

[0145] In certain embodiments, the target protein species is selected from the group consisting of a monoclonal antibody (mAb), a fusion protein, a viral capsid protein, an antibody-drug conjugate, a recombinant protein, and a plasma protein.

[0146] In certain embodiments, the aqueous sample contains multiple different molecular forms of a particular protein (e.g., a therapeutic protein (e.g., a mAb)), including a monomeric form and one or more aggregated forms (e.g., dimers and / or other multimers), and the target protein species is the monomeric form of the particular protein.

[0147] In certain embodiments, the aqueous sample contains one or more subspecies of a particular protein, each having a particular desired level and / or type of molecular conjugation (e.g., glycans, small molecule drugs, polyethylene glycol, etc.), and the target protein species is a particular one of the one or more subspecies.

[0148] In certain embodiments, one or more MIR analyzers detect MIR light [e.g., one or more wavelengths substantially within the MIR spectral range (e.g., about 5000 cm -1 ~about 500cm -1a quantum cascade laser (QCL)-based mid-IR spectrometer including one or more detectors aligned and operable to detect the beam of MIR light after its passage through and / or contact with the aqueous sample; one or more sampling optics aligned to direct and / or allow the passage of the beam of MIR light, and / or at least a portion thereof, toward one or more detectors; and one or more detectors aligned and operable to detect the beam of MIR light after its passage through and / or contact with the aqueous sample.

[0149] In certain embodiments, the one or more sampling optics include a flow cell including a detection channel through which the aqueous sample flows; and one or more detectors aligned and operable to detect the beam of MIR light exiting the detection channel after its transmission through the detection channel.

[0150] In certain embodiments, the QCL-based light source is a tunable QCL that operatively sweeps the emission frequency of the beam of MIR light through multiple frequencies over a scan range (e.g., the scan range is approximately 1700 cm -1 ~1400cm -1 For example, the scanning range is 1300 cm -1 ~1050cm -1 For example, the scanning range is at least 1200 cm -1 ~1000cm -1In one embodiment, the method includes illuminating an aqueous sample with a plurality of radiation frequencies by sweeping the radiation frequency of a beam of MIR light across a scan range of a tunable laser at each of one or more time points; and detecting, with one or more detectors, the beam of MIR light (e.g., (i) 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 radiation frequencies, thereby measuring a corresponding infrared (IR) spectrum including a plurality of values ​​each associated with, representing, and / or based on the detected output at a particular one of the plurality of radiation frequencies as a corresponding IR absorbance signal from the aqueous sample.

[0151] In certain embodiments, the MIR analyzer is an online sensor (e.g., as opposed to an offline or at-line sensor), and step (a) includes repeatedly measuring an 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 an IR absorbance spectrum from the aqueous sample in substantially real time).

[0152] In certain embodiments, the spectral data includes, for each of one or more time points, a corresponding amide band spectrum [e.g., an amide band spectrum from about 1800 to about 800 cm -1 (e.g., the scanning range is approximately 1700-1400 cm -1 For each particular wavelength of a plurality of sampled (e.g., radiation) wavelengths within a range of 100 Hz to 150 Hz (including a range of 100 Hz to 150 Hz), an associated absorption value representing the level of absorption by a portion of the aqueous sample at the particular wavelength.

[0153] In certain embodiments, step (c) includes receiving (e.g., and / or accessing), by the processor, a reference spectrum for the target protein species, the reference measured from a specific corresponding reference sample comprising a substantially isolated and / or highly purified target protein species (e.g., 75% or more pure (e.g., 90% or more pure (e.g., 95% or more pure))); and tracking the concentration of the target protein species over time by repeatedly using the reference spectrum at each of a plurality of time points to determine the concentration of the target protein species in the aqueous sample at each time point.

[0154] In certain embodiments, step (c) includes receiving (e.g., and / or accessing) by the processor reference spectra of one or more impurities, each associated with a particular impurity of interest, measured from a particular corresponding reference sample containing the impurity of interest in a substantially isolated and / or highly pure form (e.g., 75% or higher purity (e.g., 90% or higher purity (e.g., 95% or higher purity)); and repeatedly using the reference spectra of the one or more impurities at each of a plurality of time points to determine the concentration of each of the impurities of interest in the aqueous sample.

[0155] In certain embodiments, the spectral data includes, for each of the one or more time points, a corresponding amide band spectrum, and step (c) includes determining a ratio of absorbances at at least two wavenumbers within the amide band spectrum as a measure of sample purity.

[0156] In certain embodiments, step (d) comprises causing the processor to send one or more trigger signals (e.g., voltages) to a controller unit of the purification unit and / or to downstream (from the purification unit) valves.

[0157] In certain embodiments, step (d) includes one or both of the following steps: initiating collection of the target fraction of the aqueous sample based on one or more trigger signals (e.g., 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 falls below a particular threshold); e.g., a particular one of the one or more trigger signals is a digital signal that triggers the initiation of collection of the target fraction (e.g., by transitioning from a voltage level of 0 to a voltage level of 1, or vice versa)); and stopping collection of the target fraction of the aqueous sample based on the one or more trigger signals (e.g., 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 falls below a particular threshold); e.g., a particular one of the one or more trigger signals is a digital signal that triggers the cessation of collection of the target fraction (e.g., by transitioning from a voltage level of 0 to a voltage level of 1, or vice versa)).

[0158] In certain embodiments, the target protein species is a monomeric form of a specific protein (e.g., a monoclonal antibody), and the method includes, in step (c), determining values ​​over time 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 (e.g., the relative proportion (e.g., mass) of the collected monomeric form of the specific protein relative to the total collected protein) within the total collected volume of the sample exiting the purification unit; and, in step (d), stopping collection of the aqueous sample exiting the purification unit at a specific stop time based at least in part on the concentration and / or cumulative purity values ​​of the monomeric form of the specific protein.

[0159] In certain embodiments, the aqueous sample comprises (i) one or more highly aggregated forms of a specific protein and / or (ii) one or more fragmented species of a specific protein, and step (c) comprises determining the concentration(s) of the specific protein(s) and / or the concentration(s) of the one or more fragmented species over time.

[0160] In certain embodiments, the aqueous sample contains one or more additives, and the method includes, in step (c), determining concentration and / or amount values ​​of the one or more additives in the aqueous sample exiting the purification unit at one or more time points based on the spectral data; and, in step (d), using the determined concentration and / or amount values ​​of the additives to control collection of a target fraction of the aqueous sample.

[0161] In some aspects, the disclosure provides methods for preparing a biological drug formulation comprising one or more additives, the methods comprising: (a) receiving a solution comprising a purified drug substance comprising a protein species (e.g., a monoclonal antibody); (b) injecting and / or mixing one or more additives into the solution of purified drug substance over a period of time to create an in-process drug substance solution comprising the purified drug substance and the one or more additives at relative concentrations that vary over a period of time as the one or more additives are injected and / or mixed; and (c) measuring with one or more mid-infrared (MIR) analyzers, at each of one or more time points, one or more of: (i) a corresponding mid-IR absorbance spectrum from the in-process drug substance solution; and (ii) a corresponding mid-IR absorbance spectrum from a stock solution comprising at least one of the one or more additives. (d) receiving, by a processor of the computing device, spectral data corresponding to the one or more measured mid-IR absorbance spectra; (e) determining, by the processor, for each of at least some of the one or more time points based on the spectral data, corresponding values ​​of one or more sample quality metrics, the one or more sample quality metrics comprising measures of concentration and / or purity of a subset of (i) protein species and / or (ii) one or more additives; 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 additives to obtain a final drug substance having a desired protein and / or additive content and / or purity.

[0162] In certain embodiments, step (b) comprises using an ultrafiltration / diafiltration (UF / DF) system (eg, to perform buffer exchange).

[0163] In certain embodiments, the protein species is or comprises a monoclonal antibody.

[0164] In certain embodiments, the one or more additives are or include one or more surfactants {e.g., detergents; e.g., wetting agents and / or solubilizing agents [e.g., Polysorbate 20 (Tween® 20), Polysorbate 80 (Tween® 80), Poloxamers (Pluronic® F68 and F127), Triton® X-100, Brij® 30, Brij® 35, etc.]}.

[0165] In certain embodiments, the one or more additives are or include one or more bulking agents {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, cyclodextrins and derivatives, starch derivatives, HSA, BSA)}.

[0166] In certain embodiments, one or more MIR analyzers may be configured to analyze MIR light [e.g., substantially in the MIR spectral range (e.g., about 5000 cm -1 ~about 500cm -1a QCL-based light source aligned and operable to emit a beam of MIR light that includes one or more wavelengths within a range (e.g., about 2-20 microns); passing through and / or contacting at least a portion of the stock solution and / or a portion of the in-process drug substance solution [e.g., the beam of MIR light contacts the portion of the stock solution and / or a 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 portion of the stock solution and / or a portion of the in-process drug substance solution (e.g., the beam of MIR light undergoes total internal reflection and is reflected by an evanescent wave that extends into the aqueous sample]; and one or more sampling optics aligned to direct and / or allow passage of the beam of MIR light, and / or at least a portion thereof, towards one or more detectors after its passage through or contact with the portion of the stock solution and / or the portion of the in-process drug substance solution; and a quantum cascade laser (QCL) based mid-IR spectrometer including one or more detectors aligned and operable to detect the beam of MIR light after its passage through and / or contact with the portion of the stock solution and / or the portion of the in-process drug substance solution.

[0167] In certain embodiments, the one or more sampling optics include a flow cell including a detection channel through which a portion of the stock solution and / or a portion of the in-process drug substance solution flows; and one or more detectors aligned and operable to detect the beam of MIR light exiting the detection channel after its transmission through the detection channel.

[0168] In certain embodiments, the QCL-based light source is a tunable QCL that is operable to sweep the emission frequency of the beam of MIR light through multiple frequencies over a scan range (e.g., the scan range is about 1700 cm -1 ~1400cm -1 For example, the scanning range is 1300 cm-1 ~1050cm -1 For example, the scanning range is at least 1200 cm -1 ~1000cm -1 and (ii) transmitting through a detection channel through which the portion of the stock solution and / or the portion of the in-process drug substance solution flows), the method includes, at each of one or more time points, illuminating a portion of the stock solution and / or a portion of the in-process drug substance solution with a plurality of radiation frequencies by sweeping the radiation frequency of a beam of MIR light over a scan range of a tunable laser; and measuring, with one or more detectors, a corresponding infrared (IR) spectrum comprising a plurality of values ​​as a corresponding IR absorbance signal from the portion of the stock solution and / or the portion of the in-process drug substance solution, the corresponding IR spectrum comprising a plurality of values ​​each associated with, representing, and / or based on the detected output at a particular one of the plurality of radiation frequencies.

[0169] In certain embodiments, the MIR analyzer is an online sensor (e.g., as opposed to an offline or at-line sensor), and step (c) comprises repeatedly measuring the IR absorbance spectrum over time {e.g., every 20 s or less [e.g., every 10 s or less (e.g., every 5 s or less; (e.g., every second or less))]} as the one or more additives are injected and / or mixed (e.g., thereby measuring the IR absorbance spectrum from the in-process drug substance solution in substantially real time).

[0170] In certain embodiments, the spectral data includes, for each of one or more time points, a corresponding amide band spectrum [e.g., an amide band spectrum from about 1800 to about 800 cm -1range (e.g., about 1700-1400 cm -1 For each particular wavelength of a plurality of sampled (e.g., radiation) wavelengths within a range of about 1400 to about 800 cm, the spectral data includes an associated absorbance value representing the level of absorption by a portion of the aqueous sample at the particular wavelength. In certain embodiments, the spectral data includes, for each of the one or more time points, a corresponding sugar band spectrum [e.g., a sugar band spectrum ranging from about 1400 to about 800 cm]. -1 range (e.g., about 1200 to 1000 cm -1 For each particular wavelength of a plurality of sampled (e.g., radiation) wavelengths within a range of 100 Hz to 100 Hz, an associated absorption value representing the level of absorption by a portion of the aqueous sample at the particular wavelength is included.

[0171] In certain embodiments, step (e) includes receiving (e.g., and / or accessing), by the processor, a reference spectrum for the protein species, the reference measured from a specific corresponding reference sample comprising a substantially isolated and / or highly pure (e.g., 75% or more pure (e.g., 90% or more pure (e.g., 95% or more pure)) protein species; and tracking the concentration of the protein species over time by repeatedly using the reference spectrum at each of a plurality of time points to determine the concentration of the protein species in the in-process drug substance solution at each time point.

[0172] In certain embodiments, step (e) comprises receiving (e.g., and / or accessing) by the processor one or more additive reference spectra, each associated with a particular additive (of one or more additives) of interest and measured from a particular corresponding reference sample containing the particular additive of interest substantially isolated and / or highly pure (e.g., 75% or higher purity (e.g., 90% or higher purity (e.g., 95% or higher purity))); and repeatedly using the one or more additive reference spectra at each of a plurality of time points to determine the concentration of each of the one or more additives of interest in the stock solution and / or in-process drug substance solution.

[0173] In certain embodiments, step (f) includes causing the processor to send one or more trigger signals (e.g., voltages) to a controller unit (e.g., of the UF / DF system; e.g., of one or more valves).

[0174] In certain embodiments, step (f) includes initiating, by the controller unit, injection and / or mixing of one or more additives (e.g., 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 falls below a particular threshold); e.g., a particular one of the one or more trigger signals is a digital signal that triggers the initiation of injection and / or mixing (e.g., by transitioning from a voltage level of 0 to a voltage level of 1, or vice versa)) and The method includes one or both of the following steps: stopping the injection and / or mixing of the one or more additives based on the one or more trigger signals (e.g., 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 additives based on the amplitude of the analog signal (e.g., when it exceeds a particular threshold; e.g., when it falls below a particular threshold); for example, a particular one of the one or more trigger signals is a digital signal that triggers the stopping of the injection and / or mixing of the one or more additives (e.g., by transitioning from a voltage level of 0 to a voltage level of 1, or vice versa)).

[0175] In some aspects, the disclosure provides a system for obtaining a purified sample of a target protein species by mid-infrared (IR) spectroscopy-based bioprocess monitoring and control, provided the system includes one or more mid-infrared (IR) analyzers (e.g., each operable to measure a plurality of mid-IR absorbance spectra over time (e.g., based on communication with one or more signal / processors) by measuring, at each of a plurality of time points, a corresponding mid-IR absorbance spectrum from an aqueous sample exiting a purification unit, the aqueous sample comprising one or more protein species, including the target protein species); a processor of a computing device; and a memory having instructions stored thereon, the instructions, when executed by the processor, (a) receiving spectral data corresponding to a plurality of measured mid-IR absorbance spectra, each measured by one or more MIR analyzers at a corresponding one of a plurality of time points, from an aqueous sample exiting a purification unit, the aqueous sample comprising one or more protein species, including a target protein species; (b) determining, for each of at least some of the plurality of time points, based on the spectral data, a corresponding value of one or more sample quality metrics comprising a measure of the concentration and / or purity of the target protein species in the aqueous sample; and (c) using the determined value of the one or more sample quality metrics to control collection of a target fraction of the aqueous sample (e.g., during a particular collection window), thereby obtaining a purified sample of the target protein species.

[0176] In some aspects, the disclosure provides a system for preparing a biological drug formulation containing one or more additives, the system including one or more mid-infrared (MIR) analyzers; a processor of a computing device; and memory having instructions stored thereon, which, when executed by the processor, cause the processor to: (a) at each of one or more time points, generate by the one or more MIR analyzers one or more of: (i) an in-process drug substance solution, the in-process drug substance solution including a purified drug substance and one or more additives injected and / or mixed therein over time; and (ii) a stock solution including at least one of the one or more additives. (b) for each of at least some of the one or more time points, based on the spectral data, determine corresponding values ​​of one or more sample quality metrics comprising a measure of concentration and / or purity of a subset of (i) protein species and / or (ii) one or more additives; 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 additives to obtain a final drug substance having a desired protein and / or additive content and / or purity.

[0177] Features of embodiments described with respect to one aspect of the disclosure may also be applied with respect to other aspects of the disclosure.

[0178] The above and other objects, aspects, features, and advantages of the present disclosure will become more apparent and will be better understood by referring to the following description taken in conjunction with the accompanying drawings. [Brief explanation of the drawings]

[0179] [Figure 1A] FIG. 1A is a graph showing the absorption spectrum of a biological material, according to an exemplary embodiment.

[0180] [Figure 1B] FIG. 1B is a schematic diagram illustrating one particular vibration mode, according to an example embodiment.

[0181] [Figure 2-1] FIG. 2 is a graph and schematic diagram showing nucleic acid absorption in mid-IR, according to an exemplary embodiment. [Figure 2-2] FIG. 2 is a graph and schematic diagram showing nucleic acid absorption in mid-IR, according to an exemplary embodiment.

[0182] [Figure 3] FIG. 3 is a schematic diagram illustrating protein amide band vibrations and their association with secondary structure conformation, according to an exemplary embodiment.

[0183] [Figure 4] FIG. 4 is a diagram comparing mid-IR absorption spectroscopy with UV absorption, according to an example embodiment.

[0184] [Figure 5] FIG. 5 shows two graphs of UV absorption, according to an example embodiment.

[0185] [Figure 6A] FIG. 6A is a schematic diagram of FT-IR spectroscopy, according to an example embodiment.

[0186] [Figure 6B] FIG. 6B is a schematic diagram of a tunable QCL-based spectroscopy method according to an example embodiment.

[0187] [Figure 6C] FIG. 6C is a schematic diagram illustrating the operation of FT-IR spectroscopy and tunable QCL spectroscopy according to an example embodiment.

[0188] [Figure 6D]FIG. 6D is a schematic diagram illustrating the operation of FT-IR spectroscopy and tunable QCL spectroscopy according to an example embodiment.

[0189] [Figure 6E] FIG. 6E is a schematic diagram illustrating the operation of FT-IR spectroscopy and tunable QCL spectroscopy according to an example embodiment.

[0190] [Figure 6F] FIG. 6F is a schematic diagram illustrating the operation of FT-IR spectroscopy and tunable QCL spectroscopy according to an example embodiment.

[0191] [Figure 7] FIG. 7 is a graph illustrating the spectral brightness of a particular mid-IR source, according to an example embodiment.

[0192] [Figure 8] FIG. 8 is a graph showing water absorption in mid-IR, according to an exemplary embodiment.

[0193] [Figure 9A] FIG. 9A is a schematic diagram of a tunable QCL-based spectrometer according to an example embodiment.

[0194] [Figure 9B] FIG. 9B is a schematic diagram illustrating the tuning range of a spectrometer based on two QCLs, according to an example embodiment.

[0195] [Figure 10A] FIG. 10A is a schematic diagram illustrating certain steps and mathematical processes used to obtain an absorbance spectrum of an analyte present in a mobile phase, according to one or more exemplary embodiments.

[0196] [Figure 10B]FIG. 10B is a schematic diagram illustrating certain steps and mathematical processes used to obtain an absorbance spectrum of an analyte present in a mobile phase, according to one or more exemplary embodiments.

[0197] [Figure 10C] FIG. 10C is a schematic diagram illustrating certain steps and mathematical processes used to obtain an absorbance spectrum of an analyte present in a mobile phase, according to one or more exemplary embodiments.

[0198] [Figure 10D] FIG. 10D is a schematic diagram illustrating certain steps and mathematical processes used to obtain an absorbance spectrum of an analyte present in a mobile phase, according to one or more exemplary embodiments.

[0199] [Figure 10E] FIG. 10E is a schematic diagram illustrating certain steps and mathematical processes used to obtain an absorbance spectrum of an analyte present in a mobile phase, according to one or more exemplary embodiments.

[0200] [Figure 10F] FIG. 10F is a schematic diagram illustrating certain steps and mathematical processes used to obtain an absorbance spectrum of an analyte present in a mobile phase, according to one or more exemplary embodiments.

[0201] [Figure 10G] FIG. 10G is a schematic diagram illustrating certain steps and mathematical processes used to obtain an absorbance spectrum of an analyte present in a mobile phase, according to one or more exemplary embodiments.

[0202] [Figure 10H] FIG. 10H is a schematic diagram illustrating certain steps and mathematical processes used to obtain an absorbance spectrum of an analyte present in a mobile phase, according to one or more exemplary embodiments.

[0203] [Figure 11] FIG. 11 is an exemplary graph of three IR absorbance spectra showing the absorbance of a protein-nucleic acid mixture along with spectra of individual protein and nucleic acid components, according to an exemplary embodiment.

[0204] [Figure 12] FIG. 12 is a plot of IR absorbance spectra measured from a high-quality viral vector sample (e.g., having a high proportion of full capsids) and a low-quality viral vector sample (e.g., containing a large proportion of empty capsids), shown along with a difference spectrum, in accordance with an exemplary embodiment.

[0205] [Figure 13] FIG. 13 is a schematic diagram illustrating the use of an exemplary mid-IR spectrometer for measurements at various stages and for various production units in a biopharmaceutical manufacturing process, according to an exemplary embodiment.

[0206] [Figure 14] FIG. 14 is a schematic diagram showing an empty adeno-associated virus (AAV) capsid along with one loaded with a gene cassette (AAV capsid), according to an exemplary embodiment.

[0207] [Figure 15] FIG. 15 is a block flow diagram of an exemplary process for using IR absorption data measured from a viral vector sample to determine sample quality metrics and / or control a bioproduction process, according to an exemplary embodiment.

[0208] [Figure 16A] FIG. 16A is a diagram illustrating steps in an AAV vector manufacturing process, according to an exemplary embodiment.

[0209] [Figure 16B]FIG. 16B is a diagram illustrating steps in a lentiviral vector manufacturing process, according to an exemplary embodiment.

[0210] [Figure 17] FIG. 17 is a block flow diagram of an exemplary process for using reference spectra to determine sample quality metrics and / or to control a bio-production process, according to an exemplary embodiment.

[0211] [Figure 18] FIG. 18 is a block diagram of an exemplary cloud computing environment used in certain embodiments.

[0212] [Figure 19] FIG. 19 is a block diagram of an example computing device and an example mobile computing device used in certain embodiments.

[0213] [Figure 20A] FIG. 20A is a graph showing the measurement of bovine serum albumin (BSA) using mid-IR spectroscopy, according to an exemplary embodiment.

[0214] [Figure 20B] FIG. 20B is a graph showing the measurement of bovine serum albumin (BSA) using mid-IR spectroscopy, according to an exemplary embodiment.

[0215] [Figure 20C] FIG. 20C is a graph showing the measurement of bovine serum albumin (BSA) using mid-IR spectroscopy, according to an exemplary embodiment.

[0216] [Figure 20D] FIG. 20D is a graph showing the measurement of bovine serum albumin (BSA) using mid-IR spectroscopy, according to an exemplary embodiment.

[0217] [Figure 21] FIG. 21 is a graph illustrating the use of absorption spectroscopy for monitoring a chromatographic process, according to an exemplary embodiment.

[0218] [Figure 22A] FIG. 22A is a graph illustrating the use of absorption spectroscopy for monitoring a chromatographic process, according to an exemplary embodiment.

[0219] [Figure 22B] FIG. 22B is a graph illustrating the use of absorption spectroscopy for monitoring a chromatographic process, according to an exemplary embodiment.

[0220] [Figure 23A] FIG. 23A is a graph showing the use of mid-IR spectroscopy to measure protein secondary structure content, according to an exemplary embodiment.

[0221] [Figure 23B] FIG. 23B is a graph showing the use of mid-IR spectroscopy to measure protein secondary structure content, according to an exemplary embodiment.

[0222] [Figure 23C] FIG. 23C is a graph showing the use of mid-IR spectroscopy to measure protein secondary structure content, according to an exemplary embodiment.

[0223] [Figure 23D] FIG. 23D is a graph showing the use of mid-IR spectroscopy to measure protein secondary structure content, according to an exemplary embodiment.

[0224] [Figure 24] FIG. 24 is a set of graphs illustrating the use of mid-IR spectroscopy to measure protein secondary structure content, according to an illustrative embodiment.

[0225] [Figure 25]FIG. 25 illustrates the use of mid-IR absorption spectroscopy to calculate protein relative density, according to an exemplary embodiment.

[0226] [Figure 26A] FIG. 26A is a graph showing the variation of mid-IR baseline with conductivity during elution, according to an exemplary embodiment.

[0227] [Figure 26B] FIG. 26B is a graph showing the variation of mid-IR baseline with conductivity during elution, according to an exemplary embodiment.

[0228] [Figure 27] FIG. 27 is a graph illustrating the calculation of certain sample quality metrics described herein, according to certain exemplary embodiments.

[0229] [Figure 28A] FIG. 28A is a graph illustrating the calculation of certain sample quality metrics described herein, according to certain exemplary embodiments.

[0230] [Figure 28B] FIG. 28B is a graph illustrating the calculation of certain sample quality metrics described herein, according to certain exemplary embodiments.

[0231] [Figure 29A] FIG. 29A is a graph illustrating the calculation of certain sample quality metrics described herein, according to certain exemplary embodiments.

[0232] [Figure 29B] FIG. 29B is a graph illustrating the calculation of certain sample quality metrics described herein, according to certain exemplary embodiments.

[0233] [Figure 30A]FIG. 30A is a graph illustrating the calculation of certain sample quality metrics described herein, according to certain exemplary embodiments.

[0234] [Figure 30B] FIG. 30B is a graph illustrating the calculation of certain sample quality metrics described herein, according to certain exemplary embodiments.

[0235] [Figure 31] FIG. 31 is a graph illustrating the calculation of certain sample quality metrics described herein, according to certain exemplary embodiments.

[0236] [Figure 32A] FIG. 32A is a block flow diagram of a process for control of a production process based on IR absorption spectroscopy, according to an example embodiment.

[0237] [Figure 32B] FIG. 32B is an exemplary sketch showing the expected (hypothetical) variation in sample quality metrics measuring total protein content and protein aggregation, as described herein, over time during elution from an IEX chromatography column, according to an exemplary embodiment.

[0238] [Figure 33A] FIG. 33A illustrates a particular workflow and data processing approach for performing mid-IR spectroscopy measurements, according to various exemplary embodiments.

[0239] [Figure 33B] FIG. 33B illustrates a particular workflow and data processing approach for performing mid-IR spectroscopy measurements, according to various exemplary embodiments.

[0240] [Figure 33C]FIG. 33C illustrates a particular workflow and data processing approach for performing mid-IR spectroscopy measurements, according to various exemplary embodiments.

[0241] [Figure 34A] FIG. 34A is a schematic diagram illustrating control of elution, according to an exemplary embodiment.

[0242] [Figure 34B] FIG. 34B is a schematic diagram illustrating the control of elution, according to an exemplary embodiment.

[0243] [Figure 35] FIG. 35 is a set of graphs showing protein secondary structure monitoring during TFF injection, according to an exemplary embodiment.

[0244] [Figure 36] FIG. 36 is a graph showing the spectral range of a QCL system, according to an example embodiment.

[0245] [Figure 37] FIG. 37 is a graph showing the improvement in sensitivity in a mid-IR QCL system.

[0246] [Figure 38A] FIG. 38A is a graph showing the measurement of multiple analytes according to an exemplary embodiment.

[0247] [Figure 38B] FIG. 38B is a graph showing the measurement of multiple analytes according to an exemplary embodiment.

[0248] [Figure 39A] FIG. 39A is a graph showing the measurement of multiple analytes according to an exemplary embodiment.

[0249] [Figure 39B] FIG. 39B is a graph showing the measurement of multiple analytes according to an exemplary embodiment.

[0250] [Figure 40A] FIG. 40A is a graph showing the measurement of multiple analytes according to an exemplary embodiment.

[0251] [Figure 40B] FIG. 40B is a graph showing the measurement of multiple analytes according to an exemplary embodiment.

[0252] [Figure 41A] FIG. 41A is a graph showing protein secondary structure measurements, according to an exemplary embodiment.

[0253] [Figure 41B] FIG. 41B is a graph showing protein secondary structure measurements, according to an exemplary embodiment.

[0254] [Figure 41C] FIG. 41C is a graph showing protein secondary structure measurements, according to an exemplary embodiment.

[0255] [Figure 42] FIG. 42 is a graph showing protein secondary structure measurements, according to an exemplary embodiment.

[0256] [Figure 43A] FIG. 43A is a graph showing protein secondary structure measurements, according to an exemplary embodiment.

[0257] [Figure 43B] FIG. 43B is a graph showing protein secondary structure measurements, according to an exemplary embodiment.

[0258] [Figure 44] FIG. 44 is a schematic diagram of an exemplary system including multiple mid-IR analyzers and an external computer, according to an exemplary embodiment.

[0259] [Figure 45A]FIG. 45A is a graph plotting the results of certain repeatability tests, according to an exemplary embodiment.

[0260] [Figure 45B] FIG. 45B is a graph plotting the results of certain repeatability tests, according to an exemplary embodiment.

[0261] [Figure 45C] FIG. 45C is a graph plotting the results of certain repeatability tests, according to an exemplary embodiment.

[0262] [Figure 45D] FIG. 45D is a graph plotting the results of certain repeatability tests, according to an exemplary embodiment.

[0263] [Figure 45E] FIG. 45E is a graph plotting the results of certain repeatability tests, according to an exemplary embodiment.

[0264] [Figure 46A] FIG. 46A is a graph showing an IR absorbance spectrum for a DNA and protein (BSA) solution, according to an exemplary embodiment.

[0265] [Figure 46B] FIG. 46B is a graph showing IR absorbance spectra for various DNA-protein mixtures, according to an exemplary embodiment.

[0266] [Figure 46C] FIG. 46C is a diagram showing the molecular structure of a nucleotide base.

[0267] [Figure 46D] FIG. 46D is a plot of three IR absorbance spectra showing the absorbance of a protein-nucleic acid mixture along with the spectra of individual protein and nucleic acid components, according to an exemplary embodiment.

[0268] [Figure 47A] FIG. 47A is a graph of IR absorbance spectra measured for samples containing various concentrations of monoclonal antibody, according to an exemplary embodiment.

[0269] [Figure 47B] FIG. 47B is a graph of IR absorbance spectra measured for samples containing various concentrations of monoclonal antibody, according to an exemplary embodiment.

[0270] [Figure 48] FIG. 48 is a graph showing an absorbance-based chromatogram with variation in normalized spectral difference signal over time, according to an exemplary embodiment.

[0271] [Figure 49-1] FIG. 49 shows a set of graphs illustrating absorbance-based chromatograms along with the variation of normalized spectral difference signals over time for different versions, according to an exemplary embodiment. [Figure 49-2] FIG. 49 shows a set of graphs illustrating absorbance-based chromatograms along with the variation of normalized spectral difference signals over time for different versions, according to an exemplary embodiment.

[0272] [Figure 50A] FIG. 50A is a schematic diagram showing a system for protein purification, according to an exemplary embodiment.

[0273] [Figure 50B] FIG. 50B is a block flow diagram of a process for monitoring protein purification by mid-IR spectroscopy, according to an exemplary embodiment.

[0274] [Figure 50C] FIG. 50C is a block flow diagram of a process for determining component concentrations and / or sample quality metrics from measured IR spectra, according to an exemplary embodiment.

[0275] [Figure 51A] FIG. 51A is a graph of three reference spectra, according to an exemplary embodiment.

[0276] [Figure 51B] FIG. 51B is a graph of the simulated variation in integrated absorbance over time over the course of an ion exchange column run, according to an exemplary embodiment.

[0277] [Figure 51C] FIG. 51C is a graph showing extracted protein components from a simulated experiment, according to an exemplary embodiment.

[0278] [Figure 51D] FIG. 51D is a graph showing the variation in purity and yield determined from simulated chromatography runs, according to an exemplary embodiment.

[0279] [Figure 51E] FIG. 51E is a set of three graphs showing (i) extracted protein components, (ii) integrated absorbance, and (iii) purity and yield determined for a simulated chromatography run according to an exemplary embodiment.

[0280] [Figure 51F] FIG. 51F is a set of three graphs showing (i) extracted protein components, (ii) integrated absorbance, and (iii) purity and yield determined for a simulated chromatography run according to an exemplary embodiment.

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

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

[0283] [Figure 51I] FIG. 51I is a graph showing integrated absorbance measured over the course of a size exclusion chromatography run, according to an exemplary embodiment.

[0284] [Figure 51J] FIG. 51J is a graph showing spectra of extracted dimer, monomer, and dimer / monomer mixtures according to an exemplary embodiment.

[0285] [Figure 51K] FIG. 51K is a graph showing a fragment reference spectrum, according to an exemplary embodiment.

[0286] [Figure 52A] Figure 52A is a graph showing the mid-IR spectrum of certain buffer solutions.

[0287] [Figure 52B] Figure 52B is a graph showing the mid-IR spectrum of certain sugars.

[0288] [Figure 53] FIG. 53 is a schematic diagram illustrating monitoring of a UF / DF process with an inline mid-IR analyzer and an inline UV spectrometer, according to an exemplary embodiment.

[0289] [Figure 54A] FIG. 54A is a graph showing the measured BSA concentration over time as determined by mid-IR spectroscopy and UV absorbance.

[0290] [Figure 54B] FIG. 54B is a graph showing sucrose concentration over time as measured by mid-IR spectroscopy and Cedex assay.

[0291] [Figure 55A] Figure 55A is a graph showing the mid-IR spectra of polysorbate 80 (PS80) in water at various concentrations.

[0292] [Figure 55B] Figure 55B is a graph showing the mid-IR spectra of PS80 spiked into formulation buffer at various concentrations.

[0293] [Figure 56A] Figure 56A is a graph showing the mid-IR spectra of two sugars.

[0294] [Figure 56B] Figure 56B is a graph showing the mid-IR spectra of two proteins within the sugar absorption band.

[0295] [Figure 57A] Figure 57A is a graph showing the mid-IR spectra of a high molecular weight form of a monoclonal antibody spiked into a pure monomer solution at varying concentrations.

[0296] [Figure 57B] FIG. 57B is a graph showing details around the amide band region for the spectrum shown in FIG. 57A.

[0297] [Figure 57C] Figure 57C is a graph comparing IR peak metrics correlation with monomer purity as measured by size exclusion chromatography. DETAILED DESCRIPTION OF THE INVENTION

[0298] The features and advantages of the present disclosure will become more apparent from the detailed description of certain embodiments set forth below, particularly when taken in conjunction with the drawings in which like reference characters identify corresponding elements throughout, in which like reference numbers generally indicate identical, functionally similar, and / or structurally similar elements.

[0299] A specific definition In order that this disclosure may be more readily understood, certain terms are first defined below. Additional definitions for these and other terms are set forth throughout the specification.

[0300] a, an: As used herein, "a" or "an" in reference to claim features means "one or more" or "at least one."

[0301] Absorption data, absorption spectrum, absorbance data, absorbance spectrum: As used herein, the terms "absorption data," "absorption spectrum," "absorbance data," and "absorbance spectrum," as 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 represent signals produced by and / or indicative of light absorption by a sample, whether the data and / or underlying signals are obtained by transmission, attenuated total reflectance (ATR), reflectance, or other measurements. The use of the terms "absorption" and "absorbance" is not intended to be limiting with respect to a particular sampling / measurement arrangement, mode of display and / or representation, or unit system. For example, an absorption spectrum can be represented as a transmission spectrum in which absorption bands appear as negative peaks, or as an absorption spectrum in which the absorption peaks are positive, pointing upward. Absorption spectra can be represented in linear or logarithmic units. Although the term "absorbance" is used in IR spectroscopy to refer in certain cases to a unit that is directly proportional to concentration and path length (e.g., the logarithm of transmittance), its use herein is not intended to limit any method, system, processing technique, computation, etc., to being performed in any particular unit system.

[0302] Administration: As used herein, the term "administration" typically refers to the administration of a composition to a subject or system. Those of skill in the art are aware of various routes that can be used for administration to a subject, e.g., a human, under appropriate circumstances. For example, in some embodiments, administration may be intraocular, oral, parenteral, topical, etc. In certain embodiments, administration may be bronchial (e.g., by bronchial instillation), buccal, cutaneous (e.g., topical to the dermis, intradermal, interdermal, transdermal, etc.), enteral, intraarterial, intradermal, intragastric, intramedullary, intramuscular, intranasal, intraperitoneal, intrathecal, intravenous, intraventricular, intraspecific organ (e.g., intrahepatic), mucosal, nasal, oral, rectal, subcutaneous, sublingual, topical, tracheal (e.g., by intratracheal instillation), vaginal, vitreous, etc. In some embodiments, administration may include dosing that is intermittent (e.g., multiple doses separated in time) and / or periodic (e.g., individual doses separated by a common period of time) dosing. In some embodiments, administration may include continuous dosing (e.g., perfusion) over at least a selected period of time.

[0303] Affinity: As known in the art, "affinity" is a measure of the tightness with which two or more binding partners bind to one another. Those skilled in the art are aware of various assays that can be used to assess affinity, as well as appropriate controls for such assays. In some embodiments, affinity is assessed in a quantitative assay. In some embodiments, affinity is assessed across multiple concentrations (e.g., multiple concentrations of one binding partner simultaneously). In some embodiments, affinity is assessed in the presence of one or more potentially competing entities (e.g., which may be present in a relevant setting, e.g., a physiological setting). In some embodiments, affinity is assessed relative to a reference (e.g., having a known affinity above a certain threshold [see "positive control"] or having a known affinity below a certain threshold [see "negative control"]). In some embodiments, affinity can be assessed relative to a concurrent reference; in some embodiments, affinity can be assessed relative to a historical reference. Typically, when affinity is assessed relative to a reference, it is assessed under comparable conditions.

[0304] Amino acid: As used herein in its broadest sense, refers to any compound and / or substance that can be incorporated into a polypeptide chain, for example, through the formation of one or more peptide bonds. In some embodiments, an amino acid has the general structure HN-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 of the 20 standard L-amino acids commonly found in naturally occurring peptides. A "non-standard amino acid" refers to any amino acid other than the standard amino acids, regardless of whether it is synthetically prepared or obtained from a natural source. In some embodiments, amino acids, including the carboxy- and / or amino-terminal amino acids in a polypeptide, may contain structural modifications compared to the general structure above. For example, in some embodiments, an amino acid may be modified compared to the general structure by methylation, amidation, acetylation, pegylation, 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, such modifications may, for example, alter the circulating half-life of a polypeptide containing the modified amino acid compared to one that otherwise contains the same unmodified amino acid. In some embodiments, such modifications do not significantly alter the relevant activity of a polypeptide containing the modified amino acid compared to one that otherwise contains the same unmodified amino acid. As is clear from the context, in some embodiments, the term "amino acid" may be used to refer to a free amino acid; in some embodiments, it may be used to refer to an amino acid residue of a polypeptide.

[0305] Antibody, antibody polypeptide: As used herein, the terms "antibody polypeptide" or "antibody," or "antigen-binding fragment thereof," can be used interchangeably and refer to a polypeptide capable of binding to an epitope. In some embodiments, an antibody polypeptide is a full-length antibody, and in some embodiments, is less than full-length but comprises at least one binding site (comprising at least one, and preferably at least two, sequences having the structure of an antibody "variable region"). In some embodiments, the term "antibody polypeptide" encompasses any protein having a binding domain that is homologous or largely 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 an immunoglobulin binding domain, e.g., a binding domain that exhibits at least 70%, 80%, 85%, 90%, or 95% identity to a reference immunoglobulin binding domain. Included "antibody polypeptides" may have an amino acid sequence identical to that of an antibody found in a natural source. 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 using host systems, chemical synthesis, etc., or a combination thereof. Antibody polypeptides may be monoclonal or polyclonal. Antibody polypeptides may be members of any immunoglobulin class, including any of the human classes: IgG, IgM, IgA, IgD, and IgE. In certain embodiments, antibodies may be members of the IgG immunoglobulin class. As used herein, the terms "antibody polypeptide" or "characteristic portion of an antibody" are used interchangeably and refer to any derivative of an antibody that retains the ability to bind to an epitope of interest. In certain embodiments, an "antibody polypeptide" is an antibody fragment that retains at least a significant 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'), scFv, Fv, dsFv diabody, and Fd fragments.Alternatively, or in addition, an antibody fragment may comprise multiple chains linked together, for example, by disulfide bonds. In some embodiments, the antibody polypeptide may be a human antibody. In some embodiments, the antibody polypeptide may be humanized. A humanized antibody polypeptide may be a chimeric immunoglobulin, immunoglobulin chain, or antibody polypeptide (such as an Fv, Fab, Fab', F(ab')2, or other antigen-binding subsequence of an antibody) that contains minimal sequence derived from a non-human immunoglobulin. In general, humanized antibodies are human immunoglobulins (recipient antibody) in which residues from the recipient's complementarity-determining regions (CDRs) are replaced by residues from the CDRs of a non-human species (donor antibody), such as mouse, rat, or rabbit, having the desired specificity, affinity, and capacity.

[0306] Approximately: As used herein, the term "approximately" or "about" as applied to one or more values ​​of interest refers to a value similar to a stated reference value. In certain embodiments, the term "approximately" or "about" refers to a range of values ​​that are 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 or less than) the stated reference value, unless otherwise stated or otherwise clear from the context (except when such number exceeds 100% of possible values).

[0307] Backbone, Peptide Backbone: As used herein, the term "backbone," e.g., as in backbone or peptide or polypeptide, refers to the portion of a peptide or polypeptide chain that includes the links between the amino acids of the chain, but not the side chains. In other words, the backbone refers to the portion of a peptide or polypeptide that remains when the side chains are removed. In certain embodiments, the backbone is a chain that includes the carboxyl group of one amino acid linked to the amino group of the next amino acid via a peptide bond, etc. The backbone may also be referred to as a "peptide backbone." It should be understood that when the term "peptide backbone" is used, 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 protein.

[0308] Biopharmaceutical: As used herein, the term "biopharmaceutical" refers to a composition that is, or can be, produced by recombinant DNA technology, chemical synthesis, peptide synthesis, or purified and / or isolated from a natural source (such as human, animal, or microbial), and has a desired biological activity. A biopharmaceutical may be, for example, a protein, peptide, glycoprotein, polysaccharide, nucleic acid, phospholipid, a mixture of proteins or peptides, a mixture of glycoproteins, a mixture of polysaccharides, a mixture of nucleic acids, a mixture of one or more proteins, peptides, glycoproteins, polysaccharides, or nucleic acids, or a derivatized form and / or assembly of any of the above entities. In certain embodiments, a biopharmaceutical may be or include a living entity, such as a cell or tissue. The molecular weight of a biopharmaceutical may vary widely, from approximately 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 biopharmaceuticals include, but are not limited to, vaccines, blood and blood components, allergens, somatic cells, gene therapy, tissues, and recombinant therapeutic proteins. In certain embodiments, biopharmaceuticals are drugs used to treat diseases and / or medical conditions. Examples of biological drugs include, but are not limited to, natural or engineered antibodies or antigen-binding fragments thereof, and antibody-drug conjugates comprising antibodies or antigen-binding fragments thereof conjugated directly or indirectly (e.g., via a linker) to drugs of interest, such as cytotoxic drugs or toxins. In certain embodiments, biological drugs, such as gene therapy drugs, include vectors, such as adeno-associated viruses (AAV), adenoviruses, lentiviruses, and retrovirus vectors, together with (e.g., loaded with) nucleic acids, such as DNA or RNA. In certain embodiments, biopharmaceuticals are diagnostic agents used to diagnose diseases and / or medical conditions. For example, allergen patch testing utilizes biopharmaceuticals (e.g., biopharmaceuticals manufactured from natural substances) known to cause contact dermatitis.Diagnostic biopharmaceuticals may also include medical imaging agents such as proteins labeled with agents that provide a detectable signal to facilitate imaging, such as fluorescent markers, dyes, radionuclides, etc.

[0309] In vitro: As used herein, the term "in vitro" refers to events that take place in an artificial environment, e.g., in a test tube or reaction vessel, in cell culture, etc., rather than within a multicellular organism.

[0310] In vivo: As used herein, the term "in vivo" refers to events that occur within a multicellular organism, such as a human or non-human animal. In the context of cell-based systems, the term can be used to refer to events that occur within a living cell (as opposed to, for example, an in vitro system).

[0311] Peptide: As used herein, the term "peptide" refers to a typically relatively short polypeptide, e.g., 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.

[0312] Polypeptide: As used herein, refers to a polymeric chain of amino acids. In some embodiments, a polypeptide has an amino acid sequence that occurs in nature. In some embodiments, a polypeptide has an amino acid sequence that does not occur in nature. In some embodiments, a polypeptide has an amino acid sequence that is engineered in that it is designed and / or produced by the action of man. In some embodiments, a polypeptide may comprise or consist of natural amino acids, unnatural amino acids, or both. In some embodiments, a polypeptide may comprise or consist of only natural amino acids or only unnatural amino acids. In some embodiments, a polypeptide may comprise D-amino acids, L-amino acids, or both. In some embodiments, a polypeptide may comprise only D-amino acids. In some embodiments, a polypeptide may comprise only L-amino acids. In some embodiments, a polypeptide may comprise one or more pendant groups or other modifications modifying or attached to one or more amino acid side chains, for example, at the N-terminus of the polypeptide, at the C-terminus of the polypeptide, or any combination thereof. In some embodiments, such pendant groups or modifications may be selected from the group consisting of acetylation, amidation, lipidation, methylation, PEGylation, etc., and combinations thereof. In some embodiments, a polypeptide may be cyclic and / or may include a cyclic portion. In some embodiments, a polypeptide is not cyclic and / or does not include any cyclic portion. In some embodiments, a polypeptide is linear. In some embodiments, a polypeptide may 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 a related activity or structure and thus can be considered members of the same class or family of polypeptides.For each such class, the specification provides exemplary polypeptides within the class whose amino acid sequences and / or functions are known and / or known to those of skill in the art; in some embodiments, such exemplary polypeptides are reference polypeptides for the polypeptide class or family. In some embodiments, members of a polypeptide class or family exhibit significant sequence homology or identity with the reference polypeptide of that class; in some embodiments, share common sequence motifs (e.g., characteristic sequence elements), and / or share a common activity (in some embodiments, at a similar level or within a specified range) with all polypeptides within the class. For example, in some embodiments, member polypeptides exhibit an overall degree of sequence homology or identity with a reference polypeptide of at least about 30-40%, often about 50%, 60%, 70%, 80%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more, and / or contain at least one region (e.g., a conserved region which, in some embodiments, may be or may include a distinctive sequence element) that exhibits very high sequence identity, often greater than 90%, or even greater than 95%, 96%, 97%, 98%, or 99%. Such conserved regions typically encompass at least 3-4, often up to 20 or more amino acids; in some embodiments, a conserved region encompasses at least one 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 may comprise or consist of a fragment of a parent polypeptide.In some embodiments, a useful polypeptide may comprise or consist of multiple fragments that are each found in the same parent polypeptide in a different spatial arrangement compared to each other that is found in the polypeptide of interest (e.g., fragments that are directly linked in the parent may be spatially separated in the polypeptide of interest, or vice versa, and / or fragments may be present in a different order in the polypeptide of interest than in the parent), and thus the polypeptide of interest is a derivative of its parent polypeptide.

[0313] Protein: As used herein, the term "protein" refers to a polypeptide (i.e., a series of at least two amino acids linked together by a peptide bond). A protein may contain moieties other than amino acids (e.g., a glycoprotein, proteoglycan, etc.) and / or may be otherwise processed or modified. One of skill in the art will appreciate that a "protein" may refer to the entire polypeptide chain (with or without a signal sequence) produced by a cell, or a characteristic portion thereof. One of skill in the art will appreciate that a protein may sometimes comprise more than one polypeptide chain, for example, linked by one or more disulfide bonds or linked by other means. Polypeptides may contain L-amino acids, D-amino acids, or both, and may 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, proteins may comprise natural amino acids, unnatural amino acids, synthetic amino acids, and combinations thereof. The term "peptide" is generally used to refer to a polypeptide having a length of less than about 100 amino acids, less than about 50 amino acids, less than 20 amino acids, or less than 10 amino acids. In some embodiments, the protein is an antibody, an antibody fragment, a biologically active portion thereof, and / or a characteristic portion thereof.

[0314] 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., software functionality) that implements one or more specific machine learning algorithms, such as an artificial neural network (ANN), a convolutional neural network (CNN), a random forest, a decision tree, a support vector machine, etc., to determine one or more output values ​​for given input values. In some embodiments, a machine learning module implementing a machine learning technique is trained, for example, using a curated and / or manually annotated dataset. Such training can be used to determine various parameters of the machine learning algorithm implemented by the machine learning module, such as weights associated with layers in a neural network. In some embodiments, once a machine learning module has been trained to accomplish a particular task, such as determining various metrics described herein, the determined parameter values ​​are fixed, and the (e.g., unchanged, static) machine learning module is used to process new data (e.g., different from the training data) and accomplish its trained task without further updates to its parameters (e.g., the machine learning module does not receive feedback and / or updates). In some embodiments, the machine learning module may receive feedback, e.g., based on user reviews of accuracy, and such feedback may be used as additional training data to, e.g., dynamically update the machine learning module. In some embodiments, the trained machine learning module is a classification algorithm, e.g., a random forest classifier, with adjustable and / or fixed (e.g., locked) parameters. In some embodiments, two or more machine learning modules may be combined and implemented as a single module and / or a single software application. In some embodiments, two or more machine learning modules may also be implemented separately, e.g., as separate software applications.The machine learning module may be software and / or hardware. For example, the machine learning module may be implemented entirely as software, or certain functions of the ANN module may be performed by specialized hardware (e.g., by an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), etc.).

[0315] Mid-infrared, mid-IR, MIR: As used herein, the terms mid-infrared, mid-IR, and MIR are used interchangeably and refer to wavelengths within the range of approximately 5000 cm -1 ~about 500cm -1 (corresponding to a wavelength range of about 2 μm to about 20 μm) and / or in certain embodiments, about 3,000 cm -1 ~about 800cm -1 It refers to the part of the electromagnetic spectrum in the range (corresponding to wavelengths in the range from about 3 μm to about 12 μm).

[0316] Substantially: As used herein, the term "substantially" refers to the qualitative state of exhibiting the whole or near whole extent or degree of a desired characteristic or property.

[0317] Detailed Description The systems, architectures, devices, methods, and processes of the present disclosure are intended to encompass variations and adaptations developed using information derived from the embodiments described herein. Adaptations and / or modifications of the systems, architectures, devices, methods, and processes described herein can be performed as contemplated by this description.

[0318] Throughout this description, when architectures, articles, devices, methods, processes, and systems are described as having, including, or comprising particular components, or when processes and methods are described as having, including, or comprising particular steps, it is further contemplated that there are architectures, articles, devices, and systems of the invention that consist essentially of or consist of the recited components, and that there are processes and methods of the invention that consist essentially of or consist of the recited processing steps.

[0319] It should be understood that the order of steps or order for performing certain actions is immaterial so long as the invention remains functional. Moreover, two or more steps or actions may be conducted simultaneously.

[0320] The mention of any publication herein, for example, in the Background section, is not an admission that the publication serves as prior art with respect to any of the claims presented herein. The Background section is presented for purposes of clarity and is not intended to be a statement of prior art with respect to any claim.

[0321] The documents are incorporated herein by reference as if set forth. In the event of any inconsistency in the meaning of a particular term, the meaning provided in the Definitions section above shall control.

[0322] Headings are provided solely for the convenience of the reader - the presence and / or placement of headings is not intended to limit the scope of the subject matter described herein.

[0323] Described herein are methods and systems for monitoring and controlling the performance and operation of production units at various stages during a biopharmaceutical manufacturing process. In certain embodiments, the biopharmaceutical production monitoring and / or control techniques described herein utilize a mid-infrared (MIR) analyzer to measure infrared (IR) absorption signals from a liquid sample in substantially real time and generate IR absorption data, such as an IR spectrum. Liquid samples measured in this manner can serve as inputs and / or outputs for one or more production units used in the manufacture of the biopharmaceutical.

[0324] As described in further detail herein, mid-IR spectroscopy provides a powerful, non-destructive, label-free analytical technique that facilitates the rapid and accurate identification and characterization of liquid samples containing biological materials. In certain embodiments, IR spectral data is used by the biopharmaceutical production monitoring and control techniques described herein to determine sample quality metrics, which provide measures 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 data from other sensors to control and / or improve the operation of one or more production units, thereby facilitating compliance with demanding quality tolerances that may be required by regulations, and / or to produce increasingly effective and / or safer products, expand production capacity, improve efficiency, etc. A. Mid-infrared (MIR) spectroscopy

[0325] Referring to FIG. 1A, mid-IR spectroscopy can be used to obtain detailed information about the vibrational transitions of biomolecules, such as carbohydrates, lipids, nucleic acids, and proteins. As shown in FIG. 1B, mid-IR spectroscopy measures the IR absorption signal resulting from the vibrational modes of molecules. When a molecule is illuminated with mid-IR light (including a range of mid-IR frequencies / wavelengths), it may absorb that light to varying degrees at various frequencies / wavelengths. Molecules have a characteristic set of vibrational modes that depend, among other things, on their molecular structure and local environment. Thus, as shown in FIG. 1A, various molecules, such as lipids, proteins, nucleic acids, and carbohydrates, absorb light within one or more characteristic bands. Absorption in 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, and area, can be used to determine metrics that measure sample properties, such as total protein content, molecular identity, and / or heterogeneity. Ai mid-IR spectral band

[0326] In certain embodiments, one or more mid-IR spectral bands are associated with nucleic acid molecules. For example, as shown in Figure 2, one or more peaks in the mid-IR spectrum are associated with nucleic acid molecules at approximately 1220-1250 cm for the sample shown in Figure 2, respectively. -1 and 1075-1100 cm -1 and can therefore be associated with asymmetric and / or symmetric PO stretching bands that can be used to detect the presence of and / or characterize nucleic acids, such as DNA and / or RNA, in a sample. As explained in more detail herein, the IR absorption band associated with the asymmetric PO stretching vibrational mode is at about 1150 cm -1 ~Approx. 1250cm -1 (For example, about 1175 cm -1 ~Approx. 1250cm -1 range; for example, about 1200 cm -1 ~Approx. 1250cm -1 range; for example, about 1210 cm -1~Approx. 1230cm -1 The IR absorption associated with the symmetric PO stretching vibrational mode may be present / in the range of about 1000 cm -1 ~Approx. 1100cm -1 (For example, about 1050 cm -1 ~Approx. 1100cm -1 range; for example, about 1075 cm -1 ~Approx. 1100cm -1 range; for example, about 1075 cm -1 ~Approx. 1085cm -1 The spectral peaks may be present in / range between (ranges of 0.01 and 0.15) and (ranges of 0.01 and 0.15). As shown in Figure 2, in certain embodiments, features such as center frequencies, shapes, individual, and / or relative amplitudes of the peaks associated with these two (symmetric and asymmetric) PO stretching bands can be used to characterize types and / or specific conformations of nucleic acid molecules. Figure 2 illustrates the variation of these spectral peaks for several nucleic acid molecules, including, for example, single- and double-stranded DNA and RNA.

[0327] In certain embodiments, one or more mid-IR spectral bands can be associated with a protein and used to characterize the protein. For example, as shown in Figure 1A, the amide I and amide II bands can be identified at about 1500 to about 1700 cm. -1 and can be used to detect and / or characterize protein content and / or structure within a sample. For example, without wishing to be bound by any particular theory, the amide I and amide II bands are believed to correspond to vibrational modes associated with protein peptide backbone atoms. 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 proteins in a sample. In certain embodiments, additionally or alternatively, a peak at about 1250 cm -1 ~Approx. 1350cm -1Other bands associated with protein (e.g., backbone) vibrations, such as the amide III band, which is in the range of 0.1 to 0.5, can be used to determine the presence and / or content of proteins in a sample.

[0328] Additionally or alternatively, in certain embodiments, measurement of the amide I band can be used to characterize the secondary structure and / or changes thereof of one or more proteins in a sample. For example, without wishing to be bound by any particular theory, the amide I band is believed to arise from molecular vibrations (C=O stretching vibrations) in the protein peptide backbone and to be sensitive to protein secondary structure (e.g., conformation). For example, as shown in Figure 3, protein secondary structural motifs such as random coil, alpha helix, and beta sheet each result in a characteristic amide I absorption peak with a characteristic center frequency, linewidth, and splitting.

[0329] With respect to protein measurement, in certain embodiments, mid-IR spectroscopy techniques offer several advantages over ultraviolet (UV) absorption measurements. Referring to Figures 4 and 5, in particular, UV absorption measures absorption at or near 280 nm, which arises from three specific amino acids—tyrosine (Tyr), tryptophan (Trp), and phenylalanine (Phe). Therefore, not all amino acids in a protein produce measurable UV absorption signals. In contrast, all amino acids (side chains and / or peptide bonds connecting them) can contribute detectable absorption in the IR, which can be measured by mid-IR spectroscopy. For example, the absorption characteristics of amide I, amide II, and amide III are associated with the vibrational modes of protein backbone atoms and are therefore produced by all proteins and roughly correspond (in intensity, i.e., absorption level) to amino acid counts. Without wishing to be bound by any particular theory, the amide I band is believed to arise (primarily) from C=O bond vibration, while amides II and III are believed to arise from NH and CN bond vibrations (their out-of-phase and in-phase combinations, respectively). Therefore, mid-IR-based measurements are not limited to specific molecular weights and / or types of proteins. Furthermore, UV absorption is adversely affected by weak linearity and cannot be used to assess (e.g., deconvolute) the heterogeneity of protein mixtures. As described and demonstrated herein, mid-IR absorption spectroscopy, in particular, can be used to measure total protein concentration and, additionally or alternatively, to quantify the heterogeneity of protein mixtures, providing functionality and insight into protein samples that is unattainable using UV absorption measurements. A.ii Medium IR analyzer

[0330] In certain embodiments, the production monitoring and / or control techniques described herein utilize one or more mid-IR analyzers to measure infrared (IR) absorbance signals from a sample mixture and produce IR absorbance data that can be analyzed to identify and / or characterize, for example, spectral peaks associated with characteristic molecular absorption bands and / or peak characteristics. A mid-IR analyzer may be or include a system including one or more components such as a mid-IR light source, a detector, and associated lenses, e.g., (but not limited to) a sampling lens, for directing light to and / or from the sample in a specific manner and interrogating it via a specific sampling geometry. MIR light source

[0331] 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 thermal source that emits light comprising a wide range of wavelengths, e.g., with a spectral profile approximately corresponding to the blackbody spectrum at a particular temperature, selected to place a large percentage of its radiant power within the MIR.

[0332] In certain embodiments, the one or more light sources of the MIR analyzer include one or more lasers that emit light at a substantially single frequency in the MIR (e.g., within a narrow frequency band about a center wavelength). The MIR laser may be a tunable laser, such that its emission frequency can be adjusted over a specific spectral range. Examples of tunable MIR lasers include, but are not limited to, quantum cascade lasers (QCLs). Other laser-based light sources may include, but are not limited to, interband cascade lasers (ICLs), light sources based on difference frequency generation, optical frequency combs (e.g., dual-comb light sources), and the like. As described in further detail herein, tunable QCL-based light sources offer various advantageous properties, such as high spectral brightness and flexible tuning range within key spectral windows in the mid-IR. Spectral measurements

[0333] In certain embodiments, an MIR analyzer can measure IR absorption signals at multiple wavelengths and / or times to generate IR spectral data that provide a measure of the detected signal and / or absorbance at multiple wavelengths within a particular spectral range. Figures 6A-F illustrate and compare two approaches for performing IR spectral measurements. One approach is an interferometric technique called Fourier transform infrared (FT-IR) spectroscopy. As shown in Figure 6A, FT-IR spectroscopy uses a split-beam interferometer with a movable mirror to record an interference pattern as a function of the time delay between two beams. The interference pattern obtained in this manner can then be Fourier transformed to obtain an IR spectrum. As shown in Figures 6C and 6E, FT-IR spectroscopy simultaneously illuminates a sample at multiple frequencies and is therefore typically used with a broadband incoherent light source, such as the thermal source described herein.

[0334] In certain embodiments, the MIR analyzer uses a spectral scanning technique to record an IR spectrum. As shown in FIG. 6B, the spectral scanning approach can be implemented using a tunable IR laser, such as a tunable QCL. Referring to FIGS. 6D and 6F, the tunable laser emits light within a narrow frequency band at a substantially single specific emission frequency / wavelength. The emission frequency of the tunable laser can be scanned to illuminate the sample at multiple wavelengths, one at a time, within the tuning range of the tunable laser. Signals can then be detected at each of the scanned wavelengths to construct an IR spectrum, one wavelength at a time. Certain advantages of QCL-based mid-IR spectroscopy

[0335] MIR laser sources can offer advantages over thermal sources. In particular, as shown in Figure 7, laser sources such as QCLs have a thermal energy several orders of magnitude higher (approximately 10 4 ~10 6 times) spectral brightness (unit: W × sr -1 ×m -2 ×μm -1 They emit intense beams of MIR light with a power density of 100 W (e.g., watts per square meter per steradian per unit wavelength). Figure 7 shows three QCLs, each with a tuning range of approximately 2-3 μm.

[0336] Referring to FIG. 8 , in certain embodiments, the high spectral brightness provided by laser light sources such as QCLs obviates several significant drawbacks that have historically limited the application of traditional IR spectroscopy instruments, such as FTIR, that rely on thermal sources, for the measurement of biological samples and processes, particularly in aqueous environments.

[0337] First, as shown in Figure 8, liquid water has two strong, broad absorption bands in the MIR, one of which is at approximately 1638 cm -1The HOH bending mode, centered at , substantially overlaps with the amide I and amide II bands used to measure protein content and characterize structure. As a result, IR measurements in water have been limited to extremely short pathlength flow cells, which are incompatible with in-line monitoring of bioprocess workflows. Second, due to their low brightness, heat sources are typically used with cryogenically cooled (e.g., liquid nitrogen) detectors (e.g., MCT detectors) to achieve sufficient sensitivity. However, such cryogenically cooled detectors are cumbersome to operate and exhibit poor linearity, long-term stability, and reproducibility. Third, heat sources with low spectral brightness require long data acquisition times to achieve sufficient sensitivity. These long acquisition times are incompatible with the requirements of PAT.

[0338] The high spectral brightness offered by MIR laser sources, such as QCLs, overcomes these historical limitations of IR spectroscopy, enabling much longer path length measurements in water, the use of electrically cooled detectors, and rapid acquisition times. Mid-IR detector

[0339] In certain embodiments, the MIR analyzer includes one or more detectors. Various 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 may 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 cadmium mercury 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.

[0340] In certain embodiments, the MIR analyzer includes a single detector. In certain embodiments, the MIR analyzer includes two or more detectors. For example, an MIR analyzer, such as the MIR analyzer shown in FIG. 6B, may include a sample detector that detects a signal from light that has passed through the sample, and a reference detector that detects a signal from a portion of the illumination beam that has been split off (e.g., via a beam splitter) before the sample. In particular, the sample detector and reference detector can be used in this manner to offset fluctuations in laser power. Sampling Lens and Configuration

[0341] In certain embodiments, the MIR analyzer includes a sampling lens that is used to direct a beam of MIR light from an MIR light source onto and / or into a specific region of the sample and then to a detector for detecting a signal from the region of the sample. The sampling lens can be used to interrogate the sample in a specific manner, for example, but not limited to, by providing a transparent window through which the beam of light can pass, by directing the light at a specific angle onto / into the region of the sample via reflective elements such as, for example, mirrors and high refractive index materials, and by focusing the beam of light using lenses or curved (e.g., parabolic) reflectors.

[0342] For example, a sampling lens can be used to interrogate a sample using certain types of sampling arrangements, such as transmission or attenuated total reflectance (ATR) arrangements. In certain embodiments, a transmission arrangement uses a sampling lens to direct a beam of infrared light along a substantially straight path through the sample and onto a detector after passing through the sample. In this manner, the transmission arrangement measures the absorption of the IR light resulting from its propagation through the sample.

[0343] In certain embodiments, the ATR arrangement uses a sampling lens including a high-refractive index material, such as an ATR crystal, whose surface is in contact with the sample. The sampling lens directs a beam of infrared light into the high-refractive index material such that it is incident on the surface in contact with the sample at an angle above that 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, which probes the sample as an evanescent wave, rather than propagating through it. The penetration depth of the evanescent wave, and therefore its 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 certain embodiments, the ATR arrangement uses an ATR crystal with a shape that includes one or more angled surfaces through which light can enter and / or exit the ATR crystal, as well as 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 bromiodide, also known as KRS-5), and zinc selenide (ZnSe). In certain embodiments, the ATR crystal is a multi-bounce ATR crystal shaped to reflect light multiple times within the crystal to increase the effective path length used to probe the sample. In certain embodiments, the ATR crystal is located at the end of a fiber probe. In certain embodiments, the fiber probe itself can be used as the high refractive index material to implement an ATR sampling configuration. Examples of high refractive index materials used for IR fiber probes include, but are not limited to, the chalcogenide class, silver halides, and fluoride glasses, such as ZBLAN. Flow cell

[0344] In certain embodiments, the MIR analyzer includes a flow cell. According to various embodiments, a flow cell suitable for any application can be used. Non-limiting examples of MIR analyzers including flow cells are described in detail, for example, in U.S. Patent No. 10,753,856, issued August 25, 2020; U.S. Patent Application Publication No. 2021 / 0405001A1, published December 30, 2021; and U.S. Patent No. 11,119,079, issued September 14, 2021, the contents of each of which are incorporated herein by reference in their entirety. Examples of MIR spectrometers based on QCLs

[0345] Various MIR analysis devices based on QCL light sources are described in detail, for example, in U.S. Patent No. 10,753,856, issued August 25, 2020, U.S. Patent Application Publication No. 2021 / 0405001A1, published December 30, 2021, and U.S. Patent No. 11,119,079, issued September 14, 2021, the contents of each of which are incorporated by reference in their entirety into this specification.

[0346] FIG. 9A shows an example of a QCL-based MIR analyzer for performing absorption measurements in solution. The QCL-based MIR analyzer includes a tunable QCL laser source. As shown in inset 904, the QCL source is a scanning source that repeatedly sweeps its emission wavelength through a specific tuning range, completing a full spectral scan (i.e., across the entire tuning range) approximately every second. Referring to FIG. 9B, different QCL sources may have different tuning windows and can therefore be used to probe different portions of the MIR spectral range.

[0347] For example, as shown in Figure 9B, a commercially available QCL-based IR spectrometer—Culpeo-LA-P from Daylight Solution—detects a peak at approximately 1725 cm -1 ~Approx. 1375cm -1QCLs have a spectral window in the range of 1375 cm, which can be used to measure, for example, amide bands associated with proteins, and thus can be used for protein detection and characterization. Another QCL (e.g., Daylight Culpeo-LA-S) has a spectral window in the range of 1375 cm, for example, which can be used to measure amide bands associated with proteins, and therefore can be used for protein detection and characterization. -1 ~Approx. 1025cm -1 or approximately 1225cm -1 ~About 1000cm -1 The QCL sources may have different spectral scanning windows, particularly those associated with sugars, polysaccharides, and nucleic acids. In certain embodiments, multiple QCL sources can be used to cover a desired spectral range. For example, two QCL sources can be combined to scan at approximately 1725 cm. -1 ~Approx. 1025cm -1 In certain embodiments, two or more QCL sources can be contained in a single MIR analyzer such that they share at least a portion of the sampling lens and / or detector. In certain embodiments, two separate MIR analyzers, each containing a specific QCL with a specific tuning window, can be used to provide the desired spectral range.

[0348] Commercial implementations of example QCL-based IR spectrometers offer various performance characteristics advantageous for IR spectroscopy-based measurements during biological production processes. For example, QCL-based IR spectrometers may enable quantitative measurements to be performed in substantially real time, e.g., at rates of about 1 Hz, over a wide dynamic range (e.g., from less than 0.1 to greater than 300 mg / mL; e.g., from about 0.001 to greater than 300 g / L). In certain embodiments, QCL-based IR spectrometers are compatible with flow rates from up to about 10 L / min and / or can probe sample volumes as small as picoliters. In certain embodiments, QCL-based IR spectrometers are modular instruments and / or suitable for in-line and / or at-line measurements.

[0349] Table 1 below shows the advantages of the QCL-IR spectrometer system compared to other techniques for measuring biological production processes. [Table 1] A.iii Absorbance Data and Preprocessing

[0350] In certain embodiments, the IR absorbance signals measured from the sample are combined and / or mathematically pre-processed to produce an IR absorbance spectrum indicative of the absorbance of the sample. For example, Figures 10A-H show the various steps and mathematical processes used to obtain an absorbance spectrum of an analyte present in a mobile phase. In particular, as shown in Figures 10A-H, one or more reference spectra representing the mobile phase without the analyte can be measured and divided / subtracted to obtain the spectrum of the analyte present in a sample containing the mobile phase. B. Real-Time Bioproduction Monitoring and Process Control

[0351] In certain embodiments, the bioproduction monitoring and / or control techniques described herein utilize analytical techniques, particularly the mid-IR spectroscopy described herein, for the analysis of raw materials in process monitoring and control, and also for end-product analysis. In certain embodiments, the techniques described herein are implemented as part of a process analytical technology (PAT) framework, utilizing mid-IR spectroscopy as an integrated component within a bioprocessing workflow, for example, to identify sources of variability, monitor and facilitate 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 across an established design space for materials used, process parameters, manufacturing, environmental, and other conditions. For example, the techniques described herein can be used as process fingerprinting tools in bioprocessing unit operations. In certain embodiments, the techniques described herein can also or alternatively be used, for example, in biopharmaceutical forensic laboratories to identify counterfeit drugs and biosimilars.

[0352] In certain embodiments, one or more mid-IR analyzers can be used as in-line sensors embedded in a process stream to monitor sample quality characteristics in substantially real time. Mid-IR-based monitoring can be used alone and / or in combination with other measurement modalities. IR spectral data can be processed by various methods, including methods for determining specific metrics associated with certain absorption bands, and can be subjected to machine learning techniques. Thus, sample quality metrics can be monitored and / or used to control bioproduction processes. 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 process control and / or improvement. Mid-IR based monitoring of Bi sample quality metrics Sample Quality Metrics

[0353] In certain embodiments, IR absorbance data obtained from a liquid sample can be used to determine one or more sample quality metrics that characterize the properties of one or more target analytes in the sample. In certain embodiments, a sample quality metric is or includes a value and / or 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 certain embodiments, a sample quality metric characterizes the properties of one or more biological analytes, such as proteins, viruses and / or virus-like particles, nucleic acids, etc. In certain embodiments, a target analyte is a desired species of biopharmaceutical that one wishes to purify and retain, such as a specific protein, viral vector, nucleic acid, or form thereof. In certain embodiments, a target analyte is an undesired impurity, such as a portion of a mixture that one wishes to remove. Sample quality metrics may include, but are not limited to, specific properties (referred to as "critical quality attributes (CQAs)") that should be within appropriate limits, ranges, or distributions to ensure desired product quality.

[0354] A sample quality metric may be a value, such as a numeric value, that provides a measure of the content of a particular target analyte, such as a particular molecular species or form thereof, in a sample. The numeric value may be a direct measurement of the physical content, such as, for example, total mass, number, concentration, etc., or may be a value that is proportional to, indicates a relative change in, or correlates with, the physical content of a particular target analyte. In certain embodiments, a sample quality metric may be a metric that indicates a particular species or condition based on whether its value falls within one or more specific (e.g., pre-specified) ranges and / or is above or below one or more thresholds.

[0355] Content of a particular analyte and / or subspecies thereof. In certain embodiments, a sample quality metric is or includes a measure of the content of a particular target analyte, such as concentration, total mass, etc., of one or more target molecules. A sample quality metric may include, for example, a real-time measurement of total protein concentration (e.g., titer), total protein mass, etc., in a sample. Additionally or alternatively, a sample quality metric may include, for example, a real-time measurement of total nucleic acid concentration (e.g., titer), total nucleic acid mass, etc., in a sample. In certain embodiments, a sample quality metric includes a (e.g., real-time) content measure of a target analyte, such as total concentration, mass, etc., of one or more of the following: lipids, polysaccharides, etc. In certain embodiments, sample quality metrics include (e.g., real-time) content measures of assemblies of multiple molecules, such as total concentration (e.g., titer), total mass, number (e.g., individual number), etc., of viruses and / or virus-like particles, such as viral vector assemblies, including, but not limited to, adenovirus, adeno-associated virus (AAV), retrovirus (e.g., lentivirus), plant-based viruses (e.g., tobacco mosaic virus), etc.

[0356] In certain embodiments, a sample quality metric may be or include a measure of the absolute and / or relative abundance of a particular species 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 may be or include a measurement of molecular conformation and / or heterogeneity, such as protein secondary structure, aggregation, identity and relative concentrations of various protein species, conjugation (e.g., glycosylation), antibody-drug conjugate ratios, etc.

[0357] For example, in certain embodiments, a sample quality metric may be a measure of the absolute or relative content of a particular protein secondary structure motif. For example, as described in further detail herein, IR spectra can be used to identify the content of protein secondary structure motifs, such as alpha helices, beta sheets, beta turns, and disordered secondary structures. Thus, in certain embodiments, a sample quality metric may be a measure, such as a numerical value, that is proportional to or correlates with the content of a particular secondary structure motif. In certain embodiments, a sample quality metric may be, for example, a measure of the relative content between two secondary structure motifs.

[0358] For example, in certain embodiments, a sample quality metric may be or may include a measure of the absolute and / or relative abundance of a particular form of a protein that results 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 whole protein, e.g., phosphorylation, glycosylation, ubiquitination, nitrosylation, lipidation, and proteolysis.

[0359] In certain embodiments, a sample quality metric may be or include a measure of the absolute and / or relative abundance 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, triplex DNA, etc., and / or the relative proportion of any of the foregoing, e.g., relative to (e.g., compared to) the abundance of total DNA, nucleic acids, etc. In certain embodiments, a sample quality metric may be or include a measure of the absolute and / or relative abundance 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 may be or include a GC content metric, which provides a measure of the total and / or relative abundance of GC bases within a sample. For example, a GC content metric may be or include a measure of total GC content in a sample, such as concentration of GC, total mass, etc. In certain embodiments, a GC content metric may be or include a measure of relative GC content, e.g., scaled to total nucleic acid content, or including a known, intended, or estimated value, such as, e.g., strand length, that results in an average measure of GC content per nucleic acid molecule / strand (e.g., the percentage of GC bases in each nucleic acid molecule, on average, in a sample).

[0360] In certain embodiments, a sample quality metric may be or may include a value indicative of the degree of aggregation, or its absence and / or presence, in a sample. For example, a sample quality metric may be a value (e.g., a numerical value) that provides a measure of the content of a particular type of aggregate, such as monomer, dimer, trimer, multimer, etc., in a sample. For example, one sample quality metric may be a value that provides a measure of, e.g., is proportional to, and / or correlates with (e.g., increases or decreases with) the content of a particular type of protein aggregate in a sample. For example, one sample quality metric may measure the monomer content, while another may measure the content of dimers and / or higher molecular weight species (e.g., dimers, trimers, etc.) in a sample. In certain embodiments, a sample quality metric may indicate a particular species or state of aggregation based on its value compared to one or more ranges and / or thresholds. For example, in certain embodiments, as described in further detail herein, a sample quality metric may indicate a monomeric protein species if its value falls within a particular range, and may indicate the presence of aggregates (e.g., dimers and / or multimers) if its value falls outside the particular range. The particular range may be a pre-specified numerical range, may be calibrated for a particular sample or protein species, or may be determined in real time based on measurements during a particular sample processing run, such as, for example, a chromatographic elution.

[0361] Analyte Identification. In certain embodiments, a sample quality metric may be or include a value that identifies the presence or absence of a particular target analyte (e.g., or a sufficient amount thereof) in a sample. For example, in certain embodiments, a sample quality metric may be or include a Boolean value having two states (e.g., 1 or 0, true or false, etc.) that indicate whether a particular target analyte, such as a desired protein or protein species and / or an undesired impurity, is present in a sample. In certain embodiments, a sample quality metric is or includes a value that identifies one or more particular analytes in a sample. For example, a sample quality metric may be or include a value or set of values ​​that encodes the identity of one or more components in a sample, such as an alphanumeric string, a set of strings and / or alphanumeric characters, a numeric array, or a Boolean array.

[0362] In certain embodiments, sample quality metrics may be associated with gene therapy products. For example, in certain embodiments, sample quality metrics may characterize the content of viral particles, nucleic acid content, and / or mixtures or assemblies thereof. In certain embodiments, sample quality metrics may be or include measurements of virus and / or capsid titers as described herein. In certain embodiments, sample quality metrics may be or include a measure of capsid content, such as total empty capsid content, total full capsid content, or a relative measure, such as a ratio, percentage, etc., of empty capsids versus full capsids.

[0363] Certain Mid-IR Bands for Determining Sample Quality Metrics. In certain embodiments, the sample quality metrics are determined using mid-IR absorbance data, such as mid-IR spectral data. In certain embodiments, the sample quality metrics are determined using mid-IR absorbance data, such as mid-IR spectral data, at about 1600 cm -1 ~Approx. 1700cm -1 or approximately 1800 cm -1 range (e.g., about 1600 cm -1 ~Approx. 1725cm -1range; for example, about 1625 cm -1 ~Approx. 1725cm -1 range; for example, about 1630 cm -1 ~Approx. 1650cm -1 amide I region) and / or from about 1500 to about 1600 cm -1 range (e.g., about 1500 cm -1 ~Approx. 1575cm -1 range; for example, about 1500 cm -1 ~Approx. 1550cm -1 range; for example, about 1540 cm -1 ~Approx. 1560cm -1 The quality of a sample is determined using mid-IR spectral data that includes the amide II region (range of about 1000 cm). In certain embodiments, mid-IR spectral data that includes one or both of the amide I and amide II spectral regions can be used to determine one or more sample quality metrics that are indicative of and / or characterize one or more protein species within a sample. In certain embodiments, the sample quality metrics are determined using mid-IR spectral data that includes one or more of the amide I and amide II regions (range of about 1000 cm), which can be used, for example, to determine one or more sample quality metrics that are indicative of and / or characterize viruses and / or nucleic acid species (e.g., DNA, mRNA, etc.) within a sample. -1 ~Approx. 1350cm -1 In certain embodiments, the sample quality metric is determined using mid-IR spectral data including a spectral range of about 1000 cm -1 ~Approx. 1700cm -1 (For example, the maximum is approximately 1800 cm -1 ) spectral range from about 1000 cm to about 1000 cm. -1 ~Maximum 1800cm -1 Systems capable of measuring can be used to monitor combined or multi-component process streams containing both viral and / or protein (e.g., monoclonal antibody) species.

[0364] In certain embodiments, a sample quality metric is determined using multiple bands, for example, to quantify both protein and nucleic acid content within a sample. In certain embodiments, measures of protein and nucleic acid content within a sample can be combined to determine a sample quality metric that additionally or alternatively measures viral capsid content and / or full / empty capsid content and / or relative proportions.

[0365] For example, Figure 11 shows an exemplary schematic of an IR absorption spectrum of a sample containing ssDNA and protein, including spectra corresponding to the individual protein (1104) and ssDNA (1106) components, as well as a composite spectrum (1102) corresponding to the raw spectrum obtained from a sample containing both ssDNA and protein. 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, the IR absorption spectrum of a sample containing ssDNA and protein is measured using the amide I (1112) and / or amide II (4014) spectral bands. -1 ~Approx. 1350cm -1 range (for example, about 1250 cm -1 ~Approx. 1325cm -1 range; for example, about 1275 cm -1 ~Approx. 1325cm -1 range; for example, about 1280 cm -1 ~Approx. 1300cm -1 Protein content can be measured using IR absorbance data within the amide III spectral region (1116) (range).

[0366] Nucleic acid content can be quantified using IR absorbance data within the spectral range associated with vibrations of asymmetric and / or symmetric phosphate stretches (PO). For example, in certain embodiments, a sample quality metric (e.g., a nucleic acid content metric) measuring nucleic acid content is determined by measuring the asymmetric PO spectral band (1118) (e.g., approximately 1150 cm). -1 ~Approx. 1250cm -1 range, for example, about 1175 cm -1 ~Approx. 1250cm -1range; for example, about 1200 cm -1 ~Approx. 1250cm -1 range; for example, about 1210 cm -1 ~Approx. 1230cm -1 can be determined using IR absorption data (in the range of 1220 cm -1 In certain embodiments, a sample quality metric (e.g., a nucleic acid content metric) measuring nucleic acid content is determined by measuring the symmetric PO spectral band (1120) (e.g., peaking at approximately 1000 cm). -1 ~Approx. 1100cm -1 range, for example, about 1050 cm -1 ~Approx. 1100cm -1 range; for example, about 1075 cm -1 ~Approx. 1100cm -1 range; for example, about 1075 cm -1 ~Approx. 1085cm -1 can be determined using IR absorption data (in the range of 1080 cm -1 (It has been shown to peak around

[0367] In certain embodiments, specific spectral bands can be used / selected to enable independent quantification of these two contents in complex samples containing mixtures of proteins and nucleic acids. For example, as shown in FIG. 11, the absorbance in the amide I spectral region (1112) of a spectrum taken from a mixture (1102) can be attributed to both protein (1104) and nucleic acid (1106) components, while the absorbance in amide II (1114) and amide III (1116) is primarily due to protein (1104). The ssDNA spectrum (4006) shown in FIG. 11 is relatively flat / minimal in both of these (amide II and amide III) regions. Thus, in certain embodiments, the amide I and / or amide III bands can be used to quantify protein content in samples in which nucleic acids are or may be present. The asymmetric and symmetric PO4 regions (1118) and (1120), where the absorbance is primarily due to nucleic acid content, can be used to quantify nucleic acid content, for example, independent of variations in protein content. Peak Metrics

[0368] In certain embodiments, the value of one or more sample quality metrics described herein can be determined and monitored by analyzing one or more absorption bands within the mid-IR spectral data. In particular, in certain embodiments, the value of one or more peak metrics is determined for each of one or more specific absorption bands. The peak metrics aim to quantify characteristics of the absorption bands, such as the frequency location (e.g., center frequency), linewidth, intensity, etc., of the one or more absorption bands, and can be calculated by a variety of techniques.

[0369] Frequency Location Metric. In certain embodiments, a peak metric is or includes a measure of the frequency location of a particular absorption band. For example, in certain embodiments, the frequency location (e.g., center frequency) of a particular absorption band can be determined directly from the absorption spectrum by determining the frequency at which the amplitude of the particular absorption band peaks (e.g., reaches a maximum), by fitting a predefined function such as a Gaussian or Lorentzian and obtaining the center frequency from the fitted function, or by other methods. Peak ) In certain embodiments, the frequency location (e.g., center frequency) of a particular absorption band may be or may include the center-of-mass frequency (ν COM ) of a particular absorption band. COM can be computed from the absorbance spectrum (e.g., mid-IR absorbance spectrum) A(ν) according to Equation 1 below:

number

[0370] Linewidth Metrics: In certain embodiments, the one or more peak metrics may be or include a linewidth measurement, such as a full width at half maximum (FWHM), calculated by various techniques, such as a drop along one or two sides from the peak, or by a function (e.g., Gaussian, Lorentzian, etc.) fit.

[0371] Peak Intensity Metrics. In certain embodiments, a peak metric may be or may include a measure of the intensity of one or more particular absorption bands, such as peak amplitude or area under the curve (AUC). In certain embodiments, the peak amplitude of a particular band may be a value at a particular (e.g., nominal) center frequency associated with the particular band, v bandAs, i.e., A band =A(ν band ) from the absorbance spectrum A(ν). In certain embodiments, the peak amplitude of a particular band is calculated as the value at the peak frequency, v Peak As, i.e., A band = Peak (band) = A(ν Peak ) from the absorbance spectrum A(ν). In certain embodiments, a measure of the intensity of one or more specific absorption bands is determined by calculating the AUC. The AUC can be calculated for a specific absorption band by integrating the absorbance spectrum A(ν) over the limits of the specific absorption band. For example, ν min ~ν max For a particular absorption band in the range of

number

[0372] In certain embodiments, AUC is calculated for a single band. In certain embodiments, AUC is calculated for two or more bands, such as two or more adjacent bands or a region containing multiple bands that are thought to be associated with a specific target analyte of interest. In such cases, the AUC for the collection of two or more bands or the entire specific region can be calculated by using a v that specifies the limits of the two or more bands or the entire specific region. min and ν max can be calculated according to the above formula (2) using

[0373] Combinations of Peak Metrics. In certain embodiments, a sample quality metric may be or be calculated from (e.g., as a function of) a particular peak metric. For example, in certain embodiments, a sample quality metric may be a particular peak metric, such as a measure of band (frequency) position. In certain embodiments, a sample quality metric may be determined from (e.g., calculated as a function of) a combination of two or more (e.g., two; e.g., three or more; e.g., a plurality) peak metrics, which may be from the same and / or different absorption bands. For example, in certain embodiments, a sample quality metric may be determined as the sum, difference, ratio, product, or other function of two or more peak metrics. In certain embodiments, a sample quality metric can be calculated as a function of two or more peak metrics, along with other values ​​or constants, such as scaling factors, normalization constants, etc., that are known a priori and / or estimated and / or can be determined, for example, by molecular structure or other known characteristics and / or from other (e.g., orthogonal) measurement approaches, e.g., other spectroscopic techniques, prior measurements, etc. For example, in certain embodiments, a sample quality metric can be determined as a linear function or combination of one or more peak metrics, e.g., having a form as shown in Equation 3 below:

number

[0374] In certain embodiments, the sample quality metrics can be determined based on and / or using one or more reference spectra, which may be or may include one or more spectra obtained from IR absorption measurements on a reference sample.

[0375] The reference spectrum can be pre-measured and stored, for example, in a proprietary database, accessed from a public database, or measured in parallel, for example substantially simultaneously, with the various processing steps.

[0376] In certain embodiments, reference samples from which reference spectra can be obtained may include, but are not limited to, samples for which one or more particular sample quality metrics are known (e.g., as determined by other, orthogonal, e.g., more expensive and / or time-consuming methods not suitable for real-time and / or in-line analysis), of known and / or desired purity, with known individual components, etc. Reference samples can be prepared to match the sample or construct being screened for better stability, expression, or binding properties to its cognate substance.

[0377] For example, in certain embodiments, one or more reference samples of known purity, e.g., of a particular molecular species (e.g., protein, nucleic acid, viral particle) and / or form thereof (e.g., particular secondary structure, glycosylation, monomeric purity), can be obtained and their IR spectra measured. In certain embodiments, the reference sample may be (e.g., intentionally) mixed with one or more impurities, such as waste, undesired molecular forms, (e.g., known) bioprocess inputs, etc., that may be incompletely converted and / or filtered out by upstream processing, etc., and the corresponding IR spectra measured.

[0378] In certain embodiments, reference spectra can be obtained and / or modified in silico by various computer processes. For example, reference spectra can be constructed by ab initio calculations for specific molecular structures. Initial, e.g., library, reference spectra can be combined, scaled, or otherwise pre-processed, for example, according to Beer's Law, to create new, adjusted reference spectra that capture specific variations in sample parameters that may be of interest, remove baselines, reflect subband analysis, exhibit second derivative spectra, etc., and are adjusted accordingly for specific sample quality metrics and / or samples.

[0379] In certain embodiments, one or more sample quality metrics, e.g., reflecting sample purity, can be determined based on a comparison between a particular target spectrum measured from a sample of unknown properties (e.g., completely or partially) and one or more such reference spectra.

[0380] For example, in certain embodiments, the reference spectrum techniques described herein can be applied to the measurement of gene therapy products, such as viral vector samples. For example, one or more reference spectra can be obtained from a high-quality reference sample that contains a desired purity in terms of full capsid percentage (e.g., full capsid percentage above a certain threshold), monomer particles, and is free of certain impurities (e.g., host cell proteins and / or host cell nucleic acids; e.g., aggregates; e.g., fragments). In certain embodiments, for example, a high-quality viral vector sample may have a full capsid percentage above a certain threshold, such as 70%, 80%, or 90%.

[0381] An aqueous sample containing one or more viral vector species can be interrogated, for example, subsequently or in parallel, by a mid-IR analyzer described herein to obtain one or more target IR absorption spectra, which can then be compared to a high-quality reference spectrum to determine a measure of sample quality.

[0382] For example, FIG. 12 shows an exemplary absorbance plot (1200) of 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 80% or more than 80% full capsids) compared to a low-quality viral vector sample spectrum (1204) having a low content of full capsids. The difference between the two spectra is observable in the absorbance plot (1200). In certain embodiments, a difference spectrum can be determined by subtracting the target spectrum from the high-quality reference spectrum (e.g., or vice versa) to show the change in absorbance as a function of wavenumber. For example, FIG. 12 shows the corresponding difference spectrum (1220) determined by subtracting the high-quality reference spectrum (1202) from the low-quality viral vector sample spectrum (1204). In the difference spectrum (1220), effects such as a frequency shift in the amide II band—particularly a red shift (e.g., a shift to lower frequencies)—and reduced absorption in the amide III region are visible as an asymmetric lineshape (1222) and a pair of negative peaks (1224), respectively. Either or both of these properties, and / or various peak metrics calculated therefrom, can be used to determine sample quality metrics that are indicative of various characteristics of viral vector sample quality.

[0383] Comparison of the target IR absorption spectrum with one or more reference spectra can be accomplished 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, pre-processed versions thereof (e.g., derivative spectra, scaled and / or baseline-corrected spectra, etc.), and / or comparison spectra calculated therefrom. Numerical similarity measures may include, but are not limited to, correlation values, covariance values, Pearson's correlation values, overlap integrals, etc.

[0384] In certain embodiments, one or more machine learning models can be used to determine one or more sample quality metrics from IR absorbance data. In particular, by way of example, the machine learning model can be trained using various reference spectra to adjust and / or optimize weights of variable (learnable) parameters in one or more network layers. Once trained, the machine learning model can then be used for estimation, i.e., to determine metrics from new, unknown sample spectra. For example, in certain embodiments, the machine learning model can receive an IR spectrum as input and generate (e.g., by estimation) determined values ​​of one or more sample quality metrics as output. 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.

[0385] In certain embodiments, similarity scores can be determined based on, for example, correlation values, covariance values, Pearson correlation values, overlap integrals, etc., as well as machine learning-based techniques described herein. In certain embodiments, similarity scores can be generated and updated substantially in real time. For example, reference spectra of full capsid and empty capsid samples can be acquired (e.g., loaded, received, or otherwise accessed) by a mid-IR analyzer and / or a processor in communication therewith. For example, to monitor viral vector production, purification, etc. (e.g., for AAV samples), mid-IR absorption data is repeatedly obtained over time, and similarity scores can be generated in real time and displayed (e.g., for real-time quality monitoring), stored (e.g., as a quality log), and / or provided for further processing, such as for controlling various parameters of the production unit described herein. Data Pre-Processing

[0386] In certain embodiments, the IR absorbance data, such as mid-IR spectral data, used to determine one or more sample quality metrics is pre-processed data. For example, in certain embodiments, one or more pre-processing steps are performed on the mid-IR spectral data before it is used to calculate certain sample quality metrics and / or used as input to a machine learning model.

[0387] For example, in certain embodiments, the mid-IR spectral data can be pre-processed by a baseline correction technique that removes background signals, such as background absorption from the mobile phase (e.g., HOH bending modes of water), to obtain a mid-IR absorption spectrum indicative of one or more analytes in the mobile phase. This technique (baseline correction by removing background signals) is referred to as background subtraction, and involves, for example, subtracting a reference (e.g., background absorbance) spectrum A from a measured spectrum A.ref Subtract the background-corrected spectrum A bg =AA ref In certain embodiments, the reference spectrum can be calculated and / or selected based on a model, such as a mixture model, to reflect the presence and / or variations in the amount of one or more background components in the mobile phase.

[0388] For example, in certain embodiments, one or more background components are or include water molecules, such that a reference spectrum is selected or calculated to reflect the approximate intensity / relative amplitude of absorption due to water molecules present in the sample. In certain embodiments, a reference spectrum can be calculated, e.g., to reflect the replacement of water molecules by analyte molecules, e.g., as described herein. In certain embodiments, multiple reference spectra can be calculated to reflect fluctuations in a particular background molecule over time, e.g., during a particular biological and / or sample processing step.

[0389] For example, if IR spectral data is measured from an aqueous sample exiting a chromatography column, parameters such as salt content, pH, etc. can be changed, e.g., stepwise or in a gradient fashion, during the chromatographic elution so that the desired, time-varying protein absorption spectrum overlaps with a time-varying background spectrum. Variations in the background spectrum may be due to the displacement of water molecules by salt, e.g., in the case of ion-exchange chromatography, where salt concentrations are varied. Thus, in certain embodiments, multiple and / or continuously scaled reference spectra are used to reflect variations in background absorption as salt concentrations increase. In certain embodiments, the selection and / or calculation of a particular reference spectrum may involve using one or more template reference spectra in conjunction with input parameter values ​​(e.g., the salt concentration curve used by the chromatography column) and / or measured sensor values, e.g., conductivity or pH values ​​measured by a sensor.

[0390] For example, in certain embodiments, mid-IR spectral data is obtained during IEX chromatography column elution, and the conductivity can be measured using a conductivity sensor. The conductivity measured by the conductivity sensor can then be used to reflect the specific water molecule concentration as the salt concentration increases, and therefore select or calculate a specific reference spectrum where salt replaces 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 entire contents of which are incorporated herein by reference) used a conductivity sensor to perform background compensation, for example, due to salt (NaCl) gradients.

[0391] In certain embodiments, an IR spectrum can be averaged, for example, to improve its quality (e.g., signal-to-noise ratio). For example, to obtain an IR spectrum at a particular time t1, multiple IR absorption spectra can be acquired consecutively for t1 within a window of t1+δ and averaged to produce a single, averaged signal 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 for the particular time point. Incorporating multiple time points and temporal changes

[0392] In certain embodiments, a sample quality metric can be determined based on IR absorption values ​​at multiple (e.g., not necessarily only the current) time points. For example, a sample quality metric can measure temporal changes or aggregate values ​​of one or more properties of an IR absorption spectrum. In certain embodiments, for example, where a particular sample quality metric can be determined using values ​​calculated from an IR absorption spectrum corresponding to a measurement performed at a single (e.g., the current) time point, a differential or time-aggregate 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., respective values ​​corresponding to a particular time point) and then combined, for example, by calculating the difference, mean, median, variance, etc.

[0393] 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 over successive times. In certain embodiments, a time-aggregate sample quality metric can be calculated based on a running sum, mean, median, mode, variance, standard deviation, etc. over a particular time window, such as a particular number of seconds and / or a backward window of measurements, and / or in a cumulative manner aggregating measurements from an initial time point to the current time point. Various differential and / or aggregate sample quality metrics can be determined based on, for example, time differences, cumulative (over time) sums, time averages, etc. between sample quality metrics, such as absorbance values ​​at particular wavenumbers, ratios between multiple absorbance values, and various peak metrics described herein.

[0394] In certain embodiments, when calculating differential and / or time-intensive metrics, the spectra can be manipulated to emphasize particular properties and / or changes of interest. For example, in certain embodiments, normalization techniques can be used that allow for the creation of normalized spectral difference metrics that facilitate the identification of compositional changes in a sample.

[0395] With reference to Equations (4-6) below, spectra can be reference normalized to provide a normalized spectrum for a sample that is independent of the concentration of a particular composition (e.g., a single specific analyte or a composition of one or more analytes). That is, the absorbance at a particular wavenumber, ν, due to a particular composition at concentration C in the sample is: i teeth, (Equation 4) A(ν i )=A i =ε i ×C×l (where l is the path length).

[0396] This absorbance is expressed as the reference wavenumber ν i The reference absorbance A ref It can be normalized by dividing by: (Formula 5) A ref =A(νref )=ε ref ×C×l.

[0397] Thus, the normalized absorbance spectrum A / A ref and each wave number ν i and the normalized absorbance is

number

[0398] Therefore, the normalized absorbance spectrum,

number

number

number

[0399] However, when the composition changes, the normalized spectral difference (Δ) includes additional contributions beyond the baseline noise term (e.g., Δ = μ i ±χi ). Thus, by detecting whether a change in the normalized spectral difference is purely noise or includes additional factors, changes in sample composition can be determined. The composition may be a single analyte, or a mixture of different analytes and / or analyte species, forms, etc. Changes in composition may therefore arise due to the addition of new, different analytes, as well as differences in the relative proportions (e.g., ratios between) of specific analytes and / or their species. When a mixture is present, a change in concentration (e.g., as opposed to composition) refers to the overall concentration of the mixture, keeping the ratios between its components constant. As explained above, such concentration changes do not affect the normalized absorbance spectrum.

[0400] The normalized spectral difference Δ can be determined in various ways. For example, Equation 7a shows the difference at a specific wavelength. In certain embodiments, the normalized spectral difference Δ can be calculated by taking the integral over one or more specific bands, as shown in Equation 7b below. The one or more specific bands may include any bands described herein, such as amide I, amide II, amide III, asymmetric and / or symmetric PO stretching bands, as well as other bands of interest not necessarily described herein.

number

[0401] Thus, 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 certain embodiments, identifying a compositional change may include analyzing the normalized spectral difference signal to detect the occurrence of a change point using various techniques, including, but not limited to, those described in Killick R., P. Fearnhead, and I. A. Eckley. “Optimal detection of changepoints with a linear computational cost.” Journal of the American Statistical Association. Vol. 107, Number 500, 2012, pp. 1590-1598. In certain embodiments, step changes can be detected to identify a compositional change. In certain embodiments, variations in one or more statistical properties of the normalized spectral difference signal can be used to identify a compositional change, for example, based on whether they exceed a certain threshold and / or fall outside a certain window (e.g., tolerance range). In certain embodiments, multiple compositional changes can be identified, for example, due to multiple variations in analyte composition, addition of different analytes at different time points, and conformational changes (e.g., any alterations in spectral line shapes) induced by aggregation, temperature, and / or other buffers (e.g., salt gradients), among others. These concentration changes may occur on different time scales and / or alter various statistical properties in different ways, and can therefore be used to distinguish between various different mechanisms that alter sample composition.

[0402] Equations 7a and 7b show the subtraction of absorbance at time t+Δt from an earlier absorbance measurement 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 earlier data from more recent (e.g., current) data, as shown in Equations 7c and 7d below:

number

[0403] Thus, changes in the composition of any stream monitored by IR spectroscopy can be monitored using the normalized spectral difference signal. That is, changes in analyte concentration result in nominally zero variation in the difference between normalized absorbance spectra, while changes in sample composition result in changes that appear in the differential spectrum above the baseline cumulative noise. This technique can be used to monitor the composition of chromatography column effluents, the composition of retentates and / or filtration products in UF / DF processes, and the progress of chemical reactions.

[0404] In certain embodiments, integration, subtraction, etc., such as those shown in Equations 7a-7d above, can be performed on the IR absorbance spectrum and / or functions thereof, e.g., pre-processed or adjusted versions of the IR absorbance spectrum. For example, in certain embodiments, the IR absorbance spectrum may be a baseline-corrected spectrum. In certain embodiments, absolute, squared, shifted versions, etc., of the IR absorbance spectrum can be used to ensure specific (e.g., positive) indications of differential signals, such as those shown and / or time-intensive signals. B.ii. Combination with additional sensors

[0405] In certain embodiments, the process monitoring and / or control techniques described herein may include and / or use data generated by one or more additional sensors (e.g., other than a mid-IR analyzer). In certain embodiments, the one or more additional sensors include sensors for measuring parameters such as temperature, pressure, pH, conductivity, flow rate, and other scale parameters. Such sensors may 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.

[0406] In certain embodiments, the one or more additional sensors may include sensors that can also measure one or more physical, chemical, biological, or microbiological properties of one or more analytes in the sample. In certain embodiments, the one or more additional sensors can be used, for example, with one or more mid-IR analyzers to generate data used to determine one or more sample quality metrics. In certain embodiments, for example, mid-IR spectral data can be used, in conjunction with data from the one or more additional sensors, to determine one or more sample quality metrics.

[0407] For example, in certain embodiments, the additional sensor is or includes a UV absorption sensor. The UV absorption sensor may be or include any sensor operable to measure the absorption of a sample in the UV (e.g., about 200 nm to about 300 nm) spectral range. In certain embodiments, the UV absorption sensor may be or include a fixed pathlength sensor that measures the UV absorption of a sample using a fixed pathlength cell. In certain embodiments, the UV absorption sensor may be or include a gradient spectroscopy sensor, such as the CTech™ SoloVPE® and / or variations / embodiments thereof, that measures UV absorption while varying the pathlength through the sample. In particular, in certain embodiments, UV absorption measurements can be used together with mid-IR absorption measurements, e.g., as an internal check and / or complementary technique. For example, in certain embodiments, the combined and / or extended range of data acquisition can, among other things, provide internal validation and quantification of different types of data. For example, in certain embodiments, one or both of UV absorption and mid-IR absorption spectral data can be used to measure total protein concentration. In certain embodiments, labeling and tagging solutions in the mid-IR also or alternatively enhance the value of UV-based methods, as well as species identification, combating counterfeiting, and product traceability. B.iii. Real-time control of PAT systems

[0408] In certain embodiments, data analytics, machine learning, artificial intelligence, and the like can be utilized in conjunction with IR spectroscopy measurements and / or sample quality metrics determined as described herein to not only provide real-time monitoring and reporting of sample quality metrics, but also or alternatively can be used for predictive analytics whereby collected data can be analyzed in real time, also or alternatively, together with historical data sets and / or empirical models to predict future process states and / or material quality in the process stream.

[0409] 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 to improve performance in various growth and purification (filtration) steps, including, but not limited to, cell culture, virus production, clarification, concentration, diafiltration, chromatography, purification, direct flow filtration, tangential flow filtration, and tangential flow depth filtration. Additionally or alternatively, predictive analytics can be utilized during analytical development, process development, and formulation to design experiments to develop highly detailed process understanding, potential failure modes, and effect and sensitivity analyses in modeled processes. Production Unit

[0410] Techniques such as these can be used in both upstream processes, such as cell culture and harvesting, and downstream applications, including purification / filtration, formulation, and fill finish unit processes. For example, Figure 13 shows various upstream and downstream bioproduction process steps, along with characteristics that can be monitored by an MIR analyzer (1302) and / or used to control process parameters by the systems and methods described herein, among others.

[0411] For example, as shown in Figure 13, raw materials used in biopharmaceutical (e.g., protein, nucleic acid, viral vector, etc.) production can be initially tested and / or monitored, for example, as they are provided to various processing steps and production units to assess sample quality metrics related to material purity, identity, and the presence of impurities, among other things. Various inputs, outputs, and portions of a bioreactor production unit, such as a seed bioreactor and / or a production bioreactor, can be monitored using the techniques described herein to, for example, confirm the identity of the desired biomolecule being produced, determine the titer, and ensure that sufficient nutrients are present and / or provided in the bioreactor.

[0412] In certain embodiments, for example, one or more of the MIR analyzers described herein can be used in conjunction with a purification (filtration) unit, such as an alternating tangential flow filtration (ATF) system, a tangential flow depth filtration (TFDF) system, a tangential flow filtration (TFF) system, a chromatography column, direct flow or normal flow filtration, ultrafiltration, diafiltration, etc.

[0413] In certain embodiments, for example, one or more sample quality metrics and / or IR spectral data may be provided to control software and / or hardware systems and components, such as supervisory control and data acquisition (SCADA), manufacturing execution systems (MES), etc., for use in making decisions regarding changes to process control parameters and to create releases.

[0414] For example, data such as the empty / full ratio (e.g., in the context of a viral vector production process) or the presence of high molecular weight molecules such as dimers, trimers, and multimers when processing a protein 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 of raw materials in a process stream and / or variations in their amount can be adjusted. Process parameters controlled and / or adjusted as described herein can include, but are not limited to, process parameters that affect certain characteristics (e.g., process parameters that affect CQAs, referred to as "critical process parameters (CPPs)") that should be within appropriate limits, ranges, or distributions to ensure the desired product quality (i.e., CQAs).

[0415] Additionally or alternatively, data such as specific sample quality metrics characterizing protein content, secondary structure, protein aggregation, viral vector content, empty-to-full ratio, capsid aggregation, etc. in real time can be used to direct decisions such as when to stop processing, open or close valves directing sample and fraction collection, and pool different materials. For example, in certain embodiments, decisions regarding column loading can be made by monitoring breakthrough from a chromatography column. For example, in one example of control, upon detection of a breakthrough, a continuous manufacturing system can be commanded to open / close a valve, redirect the process flow to the next chromatography column in-line, and / or transition a fully loaded column to an elution and / or wash stage of the process.

[0416] The techniques 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., implementing any combination of the foregoing (e.g., IEX and HIC; e.g., IEX and SEC)).

[0417] In certain embodiments, the techniques described herein can be used 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 based on, e.g., the level of aggregation measured.

[0418] For example, in certain embodiments, during the discovery and production 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 both the total concentration (also referred to as titer) of an aqueous protein mixture as well as its degree of molecular heterogeneity over a wide dynamic range of concentrations, e.g., from approximately 1 picogram / liter to approximately 500 grams / liter. Additionally, or alternatively, it is desirable to perform such measurements under either static or dynamic flow conditions of up to 100 liters / minute or more. Controlling protein heterogeneity and chromatographic elution

[0419] In certain embodiments, sample quality metrics calculated from IR absorption spectra described herein can be used to measure different types of molecular heterogeneity. For example, in certain embodiments, a mixture can be heterogeneous in terms of having different species of free (unbound) proteins. In certain embodiments, heterogeneity can result from aggregation, with the mixture containing a population of identical proteins, each of which is free (unbound) or chemically bound to one (dimer) or more (trimers, tetramers, pentamers, etc.) identical proteins, thus forming various degrees of aggregation. Additionally or alternatively, forms of molecular heterogeneity that are desirable to measure can include populations of identical proteins with different types or degrees of molecular conjugation, such as polysaccharides, glycans, polyethylene glycols, or small molecules used for therapeutic purposes.

[0420] In certain embodiments, the techniques described herein can quantify protein heterogeneity, and can do so in several different ways. For example, in certain embodiments (e.g., as described in further detail in certain examples below), protein aggregation metrics can be determined that quantify the total or percent level of aggregation. In certain embodiments, conjugation-based metrics that quantify the total or percent level of glycosylation or small molecule conjugation can be calculated from IR spectra. In certain embodiments, molecular weights can be determined and molecular weight histograms can be displayed.

[0421] In particular, quantification of the level of aggregation is particularly important since, in certain embodiments, aggregates can reduce the efficacy and safety of the formulation.

[0422] For example, during ion exchange chromatography (IEX), a heterogeneous aqueous mixture of proteins is purposefully separated in time into its individual protein components by using a chromatography column and an ionic salt gradient. In certain embodiments, it is desirable to continuously and noninvasively monitor both the total protein concentration and the degree of heterogeneity of the eluent in real time so that the collection window of the target protein can be dynamically controlled to yield optimal results (e.g., the highest purity of the monomer). In other words, it may be desirable to accurately quantify the level of protein aggregation (high molecular weight species), which most often occurs during the end of the elution trace. Optimizing the collection window can maximize the quality and yield of the protein target while also maximizing the column's lifespan.

[0423] In certain embodiments, to maximize the yield of target protein and extend the useful life of the column, the column is loaded with progressively higher concentrations of total protein. However, such practice tends to result in the formation of more aggregates and a decrease in the separation time between the target protein (monomer) peak and the first protein aggregate peak (dimer). That is, the monomer and dimer peaks gradually overlap and therefore become more unclear, creating a challenge for determining an appropriate (e.g., optimal) target protein capture window.

[0424] In certain embodiments, the techniques 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 HCP and / or DNA removal on a direct flow filter in flow-through mode during clarification prior to chromatography.

[0425] In certain embodiments, the systems and methods described herein can be used to monitor the secondary / tertiary / quaternary structure of proteins. Secondary / tertiary / quaternary structure metrics can be used to control processes in antibody production and viral vector sample production (e.g., by monitoring the secondary / tertiary / quaternary structure of capsid proteins). In one example, changes in the secondary / tertiary / quaternary structure of protein molecules can be detected and used, for example, to shunt flow differently as an operation (e.g., based on an electrical trigger signal).

[0426] Sample quality metrics indicative of protein aggregation, secondary structure motifs, and other characteristics can be monitored by calculating one or more peak metrics from the IR absorption spectrum, for example, as described in the Examples below. For example, in certain embodiments, frequency locations, such as center frequencies and / or center-of-mass frequencies, as described herein can be determined for one or more of the amide I band, the amide II band, and the amide III band and monitored to track the level of protein aggregation. In certain embodiments, the sample quality metric is a protein aggregation metric indicative of the level of protein aggregation within the sample. In certain embodiments, the protein aggregation metric is determined based on (e.g., as) the frequency location (e.g., center frequency; e.g., center-of-mass frequency) of the amide I band. In certain embodiments, the protein aggregation metric is determined based on (e.g., as) the frequency location (e.g., center frequency; e.g., center-of-mass frequency) of the amide II band. In certain embodiments, the protein aggregation metric is determined based on (e.g., as) the frequency location (e.g., center frequency; e.g., center-of-mass frequency) of the amide III band. Monitoring and Control of Viral Vector Production

[0427] In certain embodiments, the techniques described herein can be used to monitor and control viral vector production control. In certain embodiments, the techniques described herein are suitable for use with various viral vector production techniques, including, for example, transfection-based techniques and those that utilize stable producer cell lines (e.g., for producing viral vectors or other products, such as antibodies). In particular, viral vectors are a class of macromolecules, with molecular weights exceeding 1 MDa in certain embodiments. Viral vectors can be used to infect target host cells with genetic material for purposes such as editing the genome of infected cells or directly translating specific proteins within infected cells. Thus, viral vectors used to edit the genome of infected cells can be used as gene therapy devices. In certain embodiments, viral vectors can be used to induce immune responses by directly translating specific proteins within infected cells, achieving, for example, vaccine functionality.

[0428] 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 viruses). As shown in Figure 14, viral vector capsids contain (e.g., are loaded with) gene cassettes. During production, certain viral vector particles may be empty—lacking the desired gene payload to be delivered to infected cells. Thus, in certain embodiments, it is beneficial for production control and process yield assessment to determine not only the total viral particle amount but also the percentage of intact viral particles that encapsulate the desired gene payload and are therefore suitable for the intended application.

[0429] Structurally, viral vectors are approximately 100 nm or less in diameter and comprise a (e.g., roughly spherical) protein shell that contains (e.g., encapsulates) one or more nucleic acid strands, such as DNA or RNA (e.g., mRNA), of varying lengths with respect to the number of nucleotide bases. Thus, techniques for quantifying protein and nucleic acid content in a mixture, such as those described herein, can be utilized to determine sample quality metrics such as total nucleic acid content (e.g., concentration), viral capsid content (e.g., concentration), and percentage of full capsids. In particular, as described herein, the mid-IR organic fingerprint band, spanning approximately 1,000-1,800 wavenumbers, contains multiple spectral subregions (subbands) that can be assigned to either the capsid, the genetic payload (cassette) contained within the capsid, or a combination of the capsid and genetic payload. The methods described herein allow for the full range of the mid-IR fingerprint spectral region (approximately 1,000 cm). -1 ~Approx. 1800cm -1 ), and / or its various subbands. Such methods can quantify capsid, genetic material concentrations and / or their volume ratios, and can additionally or alternatively quantify the difference between the composite spectrum of a sample and that of a purified, high-quality sample.

[0430] For example, as described herein and shown in Figure 11, various spectral bands can be used to measure protein and / or nucleic acid content in a sample, including mixtures, where certain bands, such as amide II and / or amide III, can be used to independently quantify protein content with respect to nucleic acid concentration. Because viral vectors are structurally protein capsids that encase nucleic acid material, spectral absorption within these bands can be used (e.g., in conjunction with the Beer-Lambert law) to determine protein content and nucleic acid metrics that quantify protein and nucleic acid concentrations, respectively, which can then be used to quantify (protein) capsid and (nucleic acid) payload metrics.

[0431] For example, in certain embodiments, total capsid concentration can be calculated based on (e.g., a scaled version of) a protein content metric that can be determined based on absorption in the amide I, amide II, and / or amide III regions. In certain embodiments, the use of the amide II and / or amide III spectral regions is desirable, facilitating independent quantification of protein content, because nucleic acid spectra have relatively little absorption in these (amide II and amide III) regions. The absorption intensity of 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 bands, such as a peak amplitude or area under the curve (AUC) measure. Thus, the protein content metric and / or total capsid content can be determined based on a peak intensity measure for one or both (e.g., a linear combination) of the amide II and amide III bands.

[0432] In certain embodiments, nucleic acid content metrics, such as total nucleic acid concentration, can be calculated by one or more peak metrics calculated based on the asymmetric and / or symmetric P04 bands. Absorption intensity in the asymmetric and / or symmetric P04 region can be determined using peak intensity metrics that measure the intensity of the asymmetric P04 and / or symmetric P04 bands, such as peak amplitude ("peak") or area under the curve ("AUC") measures. Thus, nucleic acid metrics can be determined based on peak intensity measures for one or both (e.g., linear combinations) of the asymmetric and / or symmetric P04 bands.

[0433] In certain embodiments, a protein content metric and / or a nucleic acid metric can be used to calculate a full capsid fraction that provides a measure (e.g., a ratio, a percentage, etc.) of the proportion of capsids that are full, i.e., successfully loaded with the desired genetic payload. 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.

[0434] In certain embodiments, other peak metrics can be used to determine sample quality metrics, such as capsid content and / or full capsid percentage. For example, in certain embodiments, frequency locations (e.g., center frequencies, center of mass frequencies) can be determined for one or more of the amide I band, the amide II band, the amide III band, the asymmetric PO4 band, and the symmetric PO4 band. In certain embodiments, the frequency locations can indicate a specific percentage of full-vacant capsids.

[0435] In certain embodiments, one or more peak metrics can additionally or alternatively be used to determine the level of capsid aggregation in a sample, for example, based on variations in frequency position. In certain embodiments, one or more protein secondary structure metrics can be determined for a viral vector sample as described herein. For example, combining ion exchange chromatography (charge separation of viral capsids) with mid-IR measurements (changes in protein secondary structure) can result in higher purity by selectively isolating full capsids.

[0436] Sample quality metrics associated with viral vector production described herein can be stored, displayed, or provided as trigger signals, for example, to monitor, interactively adjust, and / or automatically adjust process parameters. Figure 15 shows an exemplary process (1500) for real-time assessment and monitoring of viral process production by mid-IR spectroscopy. In certain embodiments, a raw IR absorption spectrum from an aqueous sample containing a viral vector species is collected (4302). In certain embodiments, various pre-processing steps, such as smoothing, decimation, baseline correction (e.g., to account for water temperature shifts, water displacement, etc.), can be performed (1504). The IR absorption spectrum can then be used to determine sample quality metrics (1506), such as protein content, nucleic acid content, viral capsid content, and full capsid percentage. In certain embodiments, sample quality metrics such as these can be CQAs and / or can be 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 (1508a) and / or stored in memory (1508c). In certain embodiments, the determined sample quality metrics, such as viral capsid content and / or full capsid percentage, can be used to produce digital and / or analog electronic trigger signals (1508b), which can be used, for example, for feedforward and / or feedback process control. For example, process flow rate, flow direction, pressure, temperature, pH, etc. can be adjusted as described herein. In certain embodiments, the trigger signals can be used to direct decisions such as when to stop processing, open or close valves directing sample and fraction collection, for example, to collect specific fractions of a sample eluting from a chromatography column containing a high titer and / or high full capsid percentage.

[0437] Referring to Figures 16A and 16B, such trigger signals can be used for various processing steps in the production of viral vectors. Figures 16A and 16B show the process flow for AAV and lentiviral vector production, respectively. Thus, viral capsid content and / or full percentage can be monitored from aqueous samples, for example, during and after the steps involved in production.

[0438] For example, in-line measurement of the full proportion of viral vectors (e.g., AAV capsids, lentiviral capsids, and others) can be used to monitor the separation of empty and full capsids during the chromatographic polishing step. Chromatographic polishing can utilize, for example, anion exchange (AEX) chromatography to separate empty capsids from full capsids. 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., implementing any combination of the foregoing (e.g., IEX and HIC; e.g., IEX and SEC)). In certain embodiments, in-line monitoring of the aqueous sample as it elutes from the chromatographic column can enable observation and / or control based on the measured viral capsid content and / or full proportion to select specific target fractions of the eluate to be retained or discarded depending on the desired target purity and / or yield.

[0439] Referring to FIG. 17 , in certain embodiments, the reference spectra described herein can be used to determine sample quality metrics. For example, as shown in FIG. 17 , in one example process (1700), raw spectral data can be converted into quantitative sample quality metrics, such as CQAs, and then displayed, stored, or used to trigger process control. As shown in FIG. 17 , a raw IR absorption spectrum from an aqueous sample containing a viral vector species is collected (1702). In certain embodiments, various pre-processing steps, such as smoothing, decimation, baseline correction (e.g., to account for water temperature shifts, water displacement, etc.), can be performed (1704). In certain embodiments, the raw spectrum (1702) can be scaled (1706), with or without pre-processing (1704), to adjust the overall amplitude, for example. In certain embodiments, the raw 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. The target spectrum can be compared to one or more reference spectra, such as high-quality viral vector spectra, as described herein, to determine one or more sample quality metrics (1708). Comparing the target spectrum to the reference spectrum may include calculating a difference spectrum, a difference in first derivative, a difference in second derivative, a correlation, a covariance, a Pearson's correlation, an overlap integral, or the like. In certain embodiments, comparing the target spectrum to the reference spectrum results in a comparison spectrum, such as a difference spectrum, that can be used to calculate peak metrics over time and, ultimately, a sample quality metric. In certain embodiments, a sample quality metric can be calculated based on the target and reference spectra without necessarily calculating a comparison spectrum, for example, by calculating a covariance, an overlap component, or the like.In certain embodiments, sample quality metrics such as these may be CQAs and / or may be used to determine one or more CQAs and / or CPPs. In certain embodiments, the sample quality metrics and / or parameters determined therefrom may be displayed (1710a) and / or stored in memory (1710c). In certain embodiments, determined sample quality metrics, such as viral capsid content and / or full capsid percentage, may be used to produce digital and / or analog electronic trigger signals (1710b), which may be used, for example, for feedforward and / or feedback process control. For example, process flow rate, flow direction, pressure, temperature, pH, etc. may be adjusted as described herein. In certain embodiments, trigger signals may be used to direct decisions such as when to stop processing, open or close valves directing sample and fraction collection, for example, to collect specific fractions of a sample eluting from a chromatography column containing a high titer and / or high full capsid percentage. Integration of MIR analyzers into production units and / or process streams

[0440] In certain embodiments, one or more mid-IR analyzers are incorporated into a production unit and / or process stream, for example, to measure liquid inputs and / or outputs from a production unit. In certain embodiments, one or more mid-IR analyzers are incorporated in-line, for example, by a split stream, whereby a portion of the input / output stream is sampled through the analyzer and then returned to the main process stream for continued processing.

[0441] In certain embodiments, the ATR sampling arrangement offers advantages for integration with pilot and / or commercial-scale manufacturing production units that rely on in / out flow through large diameter channels, ranging from 1 millimeter or greater in diameter to ¼ inch (6.35 mm) to 1 inch (25.4 mm). In certain embodiments, penetration greater than 40 microns (e.g., greater than 100 microns) may be impractical; therefore, the ability of the ATR arrangement to provide a finite path length, fixed by the evanescent wave, that is independent of the size of the channel through which the liquid flows dramatically facilitates integration with production-scale systems. B.iv. Control Signal Implementation

[0442] In certain embodiments, the processing, communication, equipment control, etc. described herein can be implemented in whole or in part by various components coupled to a production unit, a central processing system, or a remote device. For example, various processing, communication, and control steps can be performed by one or more firmware embedded on one or more specific devices (e.g., mid-IR analyzers, other, complementary sensor systems, production unit controllers, etc.), by a connected microprocessor system, and / or by software on an external computer connected, for example, directly, via Ethernet, or via a cloud-based connection. In certain embodiments, the connected computer can directly control process parameters or transmit data, process commands, control signals, etc. over one or more communication channels.

[0443] A variety of communication channels can be used, including, but not limited to, data packets shared over a communication network and / or communication port, analog signals transmitted as voltages or currents to a device capable of interpreting the signals, such as a PLC or another computer, standardized communication protocols / systems such as OPC-UA, Profibus, ModBus, or proprietary, proprietary communication channels. The communication channel may be wired or may use wireless protocols such as Bluetooth, WiFI, RF, or other technologies. In certain embodiments, data and / or commands are coded before being transmitted and / or shared and are decoded and used (e.g., to adjust process parameters) by the receiving device. C. Computer System and Network Environment

[0444] Referring to FIG. 18, an implementation of a network environment (1800) for use in providing the systems, methods, and architectures described herein is shown and described. In summary, referring now 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 computational resources. In some implementations, computational resources may include any hardware and / or software used to process data. For example, computational resources may include hardware and / or software capable of executing algorithms, computer programs, and / or computer applications. In some implementations, exemplary computational resources may include application servers and / or databases with storage and retrieval capabilities. Each resource provider (1802) can be connected to any other resource providers (1802) in the cloud computing environment (1800). In some implementations, the resource providers (1802) can be connected over a computer network (1808). Each resource provider (1802) can be connected to one or more computer devices 1804a, 1804b, 1804c (collectively, 1804) over the computer network (1808).

[0445] The cloud computing environment (1800) may include a resource manager (1806). The resource manager (1806) may be connected to resource providers (1802) and computing devices (1804) over a computer network (1808). In some implementations, the resource manager (1806) may facilitate the provision of computational resources by one or more resource providers (1802) to one or more computing devices (1804). The resource manager (1806) may receive a request for computational resources from a particular computing device (1804). The resource manager (1806) may identify one or more resource providers (1802) that can provide the computational resources requested by the computing device (1804). The resource manager (1806) may select a resource provider (1802) to provide the computational resources. The resource manager (1806) may facilitate the connection between the resource providers (1802) and a particular computing device (1804). In some implementations, the resource manager 1806 can establish a connection between a particular resource provider 1802 and a particular computing device 1804. In some implementations, the resource manager 1806 can redirect a particular computing device 1804 to a particular resource provider 1802 with requested computational resources.

[0446] 19 illustrates examples of a computing device (1900) and a mobile computing device (1950) that can be used to implement the techniques described in this disclosure. The computing device (1900) is intended to represent various forms of digital computers, such as laptops, desktops, 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, mobile phones, smartphones, and other similar computing devices. The components, their connections and relationships, and their functions shown here are merely examples and are not meant to be limiting.

[0447] The computer 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). The processor (1902), the memory (1904), the storage device (1906), the high-speed interface (1908), the high-speed expansion port (1910), and the low-speed interface (1912) are each interconnected using various buses, which may be mounted on a common motherboard or in other manners as needed. The processor (1902) processes instructions for execution within the computing device (1900), including instructions stored in memory (1904) or on a storage device (1906), and can display graphical information for a GUI on an external input / output device, such as a display (1916) coupled to a high-speed interface (1908). In other implementations, multiple processors and / or multiple buses can be used, along with multiple memories and memory types, as needed. Also, multiple computing devices can be connected (e.g., as a server bank, a cluster of blade servers, or a multiprocessor system), with each device providing portions of the necessary operations. Thus, as the term is used herein, when multiple functions are described as being performed by a "processor," this encompasses embodiments in which the multiple functions are performed by any number of processor(s) of any number of computing device(s). Furthermore, when a function is described as being performed by a "processor," this encompasses embodiments in which the function is performed by any number of processor(s) of any number of computing device(s) (e.g., in a distributed computing system).

[0448] The memory (1904) stores information within the computer device (1900). In some implementations, the memory (1904) is a volatile memory unit or units. In some implementations, the memory (1904) is a non-volatile memory unit or units. The memory (1904) may also be another form of computer-readable medium, such as a magnetic or optical disk.

[0449] The storage device (1906) can provide mass storage for the computer device (1900). In some implementations, the storage device (1906) can be or contain a computer-readable medium, such as a floppy disk device, a hard disk device, an optical disk device, or an array of devices, including a tape device, a flash memory or other similar solid-state memory device, or a device in a storage area network or other configuration. Instructions can be stored in an 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 memory on the processor (1902)).

[0450] The high-speed interface (1908) manages bandwidth-intensive operations for the computing device (1900), while the low-speed interface (1912) manages less bandwidth-intensive operations. Such allocation of functions is merely exemplary. In some implementations, the high-speed interface (1908) is coupled to memory (1904), a display (1916) (e.g., via a graphics processor or accelerator), and a high-speed expansion port (1910) that can accept various expansion cards (not shown). In such implementations, 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 may include various communication ports (e.g., USB, Bluetooth, Ethernet, wireless Ethernet), may be coupled to one or more input / output devices, such as a keyboard, pointing device, scanner, etc., or to a network device, such as a switch or router, for example, via a network adapter.

[0451] The computing device (1900) can be implemented in several different forms, as shown in the figures. For example, it can be implemented as a standard server (1920) or multiple times in a cluster of such servers. It can also 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 such device may contain one or more computing devices (1900) and mobile computing devices (1950), and the entire system may be made up of multiple computing devices communicating with each other.

[0452] The mobile computing device (1950) includes a processor (1952), memory (1964), and input / output devices such as a display (1954), a communication interface (1966), and a transceiver (1968), among other components. The mobile computing device (1950) may also be provided with a storage device, such as a microdrive or other device, to provide additional storage. The processor (1952), memory (1964), display (1954), communication interface (1966), and transceiver (1968) are each interconnected using various buses, and several components may be mounted on a common motherboard or in other manners as needed.

[0453] The processor 1952 can execute instructions within the mobile computing device 1950, including instructions stored in the memory 1964. The processor 1952 can be implemented as a chipset of chips including separate and multiple analog and digital processors. The processor 1952 can provide, for example, control of a user interface, applications run by the mobile computing device 1950, and coordination of other components of the mobile computing device 1950, such as wireless communication by the mobile computing device 1950.

[0454] The processor (1952) can communicate with a user via 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 electrical circuitry for driving the display (1954) to present graphical and other information to the user. The control interface (1958) can receive commands from the user and translate them for submission to the processor (1952). Additionally, an external interface (1962) can provide communication with the processor (1952) to enable short-range communication between the mobile computing device (1950) and other devices. The external interface (1962) can provide, for example, wired communication in some implementations or wireless communication in other implementations, and multiple interfaces can also be used.

[0455] The memory (1964) stores information within the mobile computing device (1950). The memory (1964) can be implemented as one or more computer-readable media or media, a volatile memory unit or units, or a non-volatile memory unit or units. An expansion memory (1974) can also be provided and connected to the mobile computing device (1950) via an expansion interface (1972), which may include, for example, a SIMM (single in-line memory module) card interface. The expansion memory (1974) can provide additional storage capacity for the mobile computing device (1950) or can store applications or other information for the mobile computing device (1950). Specifically, the expansion memory (1974) can include instructions for performing or complementing the processes described above and can also include secure information. Thus, for example, the expansion memory (1974) can be provided as a security module for the mobile computing device (1950) and can be programmed with instructions that enable secure use of the mobile computing device (1950). Additionally, secure applications can be provided via SIMM cards along with additional information, such as placing identifying information on the SIMM card in an unhackable manner.

[0456] The memory may include, for example, flash memory and / or NVRAM memory (non-volatile random access memory), as discussed below. In some implementations, the instructions are stored in an information carrier. The instructions, when executed by one or more processing devices (e.g., processor (1952)), perform one or more methods, such as those described above. The instructions may also be stored by one or more storage devices, such as one or more computer- or machine-readable media (e.g., memory (1964), expansion memory (1974), or memory on the processor (1952)). In some implementations, the instructions may be received, for example, in a propagated signal on the transceiver (1968) or external interface (1962).

[0457] The mobile computing device (1950) can communicate wirelessly via a communications interface (1966), which may include digital signal processing circuitry, if necessary. The communications interface (1966) can provide communications under various schemes or protocols, such as GSM (Global System for Mobile communications), voice calls (SMS), EMS (Enhanced Message Service), or MMS (Multimedia Messaging 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), among others. Such communications may occur via a transceiver (1968), for example, using radio frequencies. Additionally, short-range communications may occur, such as using Bluetooth, Wi-Fi, or other such transceivers (not shown). Additionally, a GPS (Global Positioning System) receiver module (1970) may provide additional navigation and location-related radio data to the mobile computing device (1950) as needed, which may be used by applications running on the mobile computing device (1950).

[0458] The mobile computing device (1950) can also communicate audibly using a voice codec (1960), which can receive spoken information from a user and convert it into usable digital information. The voice codec (1960) can also generate audible sounds for the user, such as through a speaker in the handset of the mobile computing device (1950). Such sounds may include sounds from a voice call, recorded sounds (e.g., voice messages, music files, etc.), and sounds generated by applications running on the mobile computing device (1950).

[0459] The mobile computing device (1950) can be implemented in several different forms, as shown in the figures. For example, it can be implemented as a mobile phone (1980). It can also be implemented as part of a smartphone (1982), personal digital assistant, or other similar mobile device.

[0460] Various implementations of the systems and techniques described herein can be realized in digital electronic circuitry, integrated circuits, specially designed ASICs (application-specific integrated circuits), computer hardware, firmware, software, and / or combinations thereof. These various implementations may include implementation in one or more computer programs executable and / or interpretable on a programmable system that includes at least one programmable processor, which may be special or general purpose, coupled to receive data and instructions from, and transmit data and instructions to, a storage system, at least one input device, and at least one output device.

[0461] Activities associated with implementing the system may be performed by one or more programmable processors executing one or more computer programs. All or a portion of the system may be implemented as special purpose logic circuitry, such as a field programmable gate array (FPGA) or an application specific integrated circuit (ASIC), or both. All or a portion of the system may also be implemented as special purpose 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).

[0462] These computer programs (also known as programs, software, software applications, or code) contain machine instructions for a programmable processor and may be implemented in high-level procedural and / or object-oriented programming languages ​​and / or assembly / machine languages. As used herein, the terms machine-readable medium and computer-readable medium refer to any computer program product, apparatus, and / or device (e.g., magnetic disk, optical disk, memory, programmable logic device (PLD)) used to provide machine instructions and / or data to a programmable processor, including machine-readable media that receive machine instructions as machine-readable signals. The term machine-readable signal refers to any signal used to provide machine instructions and / or data to a programmable processor.

[0463] To provide for user interaction, the systems and techniques described herein can be implemented on a computer that has 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 pointing device (e.g., a mouse or trackball) by which the user can provide input to the computer. Other types of devices can be used to provide for user interaction as well; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); input from the user can be received in any form, including acoustic, speech, or tactile input.

[0464] The systems and techniques described herein can be implemented in a computer system that includes a back-end component (e.g., as a data server), or includes a middleware component (e.g., as an application server), or includes a front-end component (e.g., a client computer with a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described herein), or includes any combination of such back-end, middleware, or front-end 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.

[0465] The computer system may include clients and servers. Clients and servers are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other.

[0466] In some implementations, the modules described herein may be separated, combined, or incorporated into single or combined modules. The modules depicted in the figures are not intended to limit the systems described herein to the software architectures shown therein.

[0467] Elements of different implementations described herein may be combined to form other implementations not specifically described above. Elements may be omitted from a process, computer program, database, etc. described herein without adversely affecting its operation. Furthermore, the logic flow depicted in the figures need not indicate a particular order or sequential order to achieve desirable results. Various separate elements may be combined with one or more individual elements to perform the functions described herein. [Example]

[0468] E. Working Example Example 1 Linearity over a wide dynamic range and at high concentrations This example demonstrates the ability of a QCL-based IR spectrometer according to certain embodiments described herein to measure protein concentration in aqueous solution over a wide dynamic range and with a high degree of linearity. Figures 20A-D show measurements of IR spectra from several different protein mixtures, each containing a specific concentration of bovine serum albumin (BSA), along with graphs comparing the concentration / associated peak metrics measured by the QCL with a priori known concentrations. Figure 20A shows several IR absorbance spectra measured for samples with different BSA concentrations. Figure 20B shows the minimum frequency (1360 cm) of the protein absorption region by calculating the AUC. -1 ~ approx. 1700cm -1 Figure 20B shows the measured concentration values ​​determined by integrating each spectrum of . The measured concentrations were determined by scaling the determined AUC and are plotted relative to the nominal reference concentration in Figure 20B. Figures 20C and 20D show similar measurements, with Figure 20D plotting the AUC directly versus the nominal BSA concentration, demonstrating a highly linear correlation, R 2 The parameters of the linear fit are shown along with the values. Thus, as shown in the graphs of Figures 20B and 20D, the measured concentrations closely match the nominal concentrations of the prepared samples and are highly linear, even at high concentrations. Example 2 Analytical chromatographic measurements and comparison with UV absorption

[0469] This example demonstrates the use of a mid-IR QCL-based spectrometer to record absorbance spectra in substantially real time to monitor the protein content of the mobile phase exiting a chromatography column during elution. Figure 21 shows the recording of mid-IR spectra over time during elution. Figures 22A and 22B compare protein content measured by UV absorbance (Figure 22A) with protein content measured by mid-IR absorbance (Figure 22B, AUC calculated as absorbance integrated over the amide I and amide II bands, showing variation over time). The mid-IR data sample quality metric, i.e., AUC, graphed in Figure 22B, exhibits lower temporal variance than the UV data, providing more accurate elution monitoring and control. Example 3 Analysis of amide subbands

[0470] This example demonstrates the use of a mid-IR QCL-based spectrometer to record amide band spectra of protein mixtures in aqueous solution and monitor changes in protein secondary structure as a function of temperature.

[0471] Figure 23A shows the IR spectra of a 10 mg / mL BSA solution measured during flow at a rate of 400 μL / min as the temperature was changed (each curve was obtained at a different time / temperature). Figure 23B shows the second derivative spectra determined from each spectrum shown in Figure 23A. Second derivative spectra can be used to highlight / monitor fluctuations in protein secondary structure, as reflected by changes in the subpeaks that make up the amide band region. Figures 23C and 23D show the fluctuations in secondary structure components, such as alpha helix, turn, beta sheet, and denatured region content, as the temperature was changed. In Figures 23C and 23D, the content of specific secondary structure motifs was measured by determining the absorbance at specific, nominal frequencies associated with the specific secondary structure motif. For example, as shown in Figure 23C, the absorbance at 1618 cm -1 The beta-sheet content was measured using the absorbance at 1692 cm -1 Measure the turnover content using absorbance at 1656 cm -1 Alpha helix content was measured using absorbance at 1645 cm -1 The absorbance at was used as a measure of the content of the variable region.

[0472] Another approach for monitoring protein secondary content is shown in Figure 24, which shows heat maps displaying amide sub-band intensities as they change with increasing temperature from room temperature to about 75°C for five different proteins. Rather than monitoring intensities at fixed nominal band positions, the amide band spectra for each of the five proteins were deconvoluted to calculate each heat map, allowing the specific peak frequencies and intensities of the component sub-bands (e.g., representing various secondary structure motifs) to be determined directly from the spectra recorded at each temperature for each of the five proteins shown in Figure 24.

[0473] In particular, tracking the location of subbands associated with secondary structural motifs allows for the observation and characterization of protein stability in different solutions / formulations. For example, a comparison of the two right-most heat maps shows the difference in temperature stability for hen egg white lysosomal (HEWL) protein in phosphate buffered saline (PBS) versus acetate buffer. For example, HEWL in PBS exhibits a single, relatively stable dominant peak over a wide range of low temperatures, while HEWL in the acetate map exhibits a meandering band at even lower temperatures, indicating lower stability.

[0474] Thus, this example demonstrates the ability to monitor protein secondary structure fluctuations in essentially real time. Example 4 Protein biophysical properties based on isosbestic points

[0475] This example demonstrates how a sample quality metric characterizing protein density can be calculated using IR absorbance spectra across and just above the amide band region.

[0476] Figure 25 shows the spectra of BSA at various concentrations. As described herein, the amide band spectrum can be used to determine various information about the protein under test. The inset shows the amide band at 1700 cm -1 25 shows a zoomed in view of the spectrum near σ. Without wishing to be bound by any particular theory, the absorbance at a particular wavenumber of a protein in a water sample compared to (i.e., normalized to) a background spectrum of pure water can be calculated as shown in FIG. 25, where A is the absorbance at a particular wavenumber and σ p is the absorption cross section of the protein, γ is the relative density of the protein to water, and σ w is the absorption cross section of water, and C p (t1) is the protein concentration (at the measured time) and L is the path length.

[0477] Protein absorption cross section σ pis about 1695cm -1 It is approximately 0 for larger wavenumbers. Therefore, 1695 cm -1 Above, A=-(γσ w )·C p (t1) L. Protein still affects absorption, not because it absorbs light itself, but instead because it displaces water. σ w is known, so the physical constants and C p (t1)·L can be measured (e.g., by integrating over the amide I and II regions) and γ—the relative (to water) protein density—can be calculated.

[0478] Thus, this example demonstrates how a measure of protein density (relative to water) can be calculated from mid-IR spectra. In certain embodiments, other measures such as hydrodynamic radius and diffusion constant can be calculated. Example 5 Baseline correction during elution using a conductivity sensor

[0479] This example shows baseline shifts due to the increasing presence of ions in measurements recorded during elution of a chromatographic column. Figure 26A shows the spectral shift before and after elution. Figure 26B shows the shift in absorbance with conductivity.

[0480] Without wishing to be bound by any particular theory, it is believed that the relatively flat shape of the water absorbance curve shown in Figure 26A makes the position of the amide II band, as calculated by, for example, the center of mass frequency, insensitive to baseline fluctuations during elution. This therefore allows, in certain embodiments, the amide II band to be used to measure changes in protein structure without the need to precisely correct for baseline fluctuations during elution. Example 6 Protein aggregation metrics from IR absorption spectra

[0481] This example illustrates the calculation of a sample quality metric that provides a measure of protein aggregation, according to various embodiments described herein.

[0482] The absorption spectrum of a static or fluid aqueous mixture of proteins is measured at approximately 800 cm -1 ~Approx. 1800cm -1 or any sub-range thereof, continuous or non-continuous. Such absorption spectra can be obtained using a variety of infrared absorption spectrometers, including, for example, Fourier transform infrared (FT-IR), quantum cascade laser infrared (QCL-IR) analyzers, etc.

[0483] Figure 27 shows a typical IR absorption spectrum of a protein measured in an aqueous buffer. The absorption spectrum in Figure 27 was calculated using a reference background taken from a nominally zero protein aqueous buffer. Protein infrared absorption spectra such as those shown in Figure 27 can be analyzed by one or more techniques described in this example to determine sample quality metrics indicative of protein heterogeneity, particularly aggregation (protein aggregation metrics) and total protein content. These sample quality metrics can then be used to generate digital or analog signals that can be used for dynamic process control and / or process optimization.

[0484] 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 approximate range values ​​in the mid-IR for the amide I and / or amide II spectral bands.

[0485] 28A and 28B show the specific peak metrics, i.e., from about 1500 to about 1600 cm -1 The "center of mass" (ν) of the amide II band roughly corresponds to the region COM Amide-II) is calculated as the center of mass of the amide II band. Figure 28A shows an example spectrum of a zero aggregation state (e.g., all monomer). Figure 28B shows a spectrum of a 5% by mass aggregation state, demonstrating that the center of mass of the amide II band varies proportionally with the relative concentration of aggregates. The nominal amide II center of mass of a highly pure protein (e.g., a monoclonal antibody) is approximately 1546.2 cm -1 (Figure 28A), but protein aggregates or fragments may be located at a distance of plus or minus 0.5 cm from this nominal center of mass. -1 It can be high or low (Figure 28B). Such small frequency shifts can be easily observed using high performance QCL-based infrared-based mid-IR analyzers. [Table 2]

[0486] In particular, the amide II center of mass frequency has been found to be relatively insensitive to baseline fluctuations during elution and can serve as a useful metric even in the absence of accurate baseline correction. Furthermore, the amide II center of mass frequency has been found to be relatively stable within a small range for measuring monomeric protein species, but will deviate from that stable range in the presence of aggregated proteins, such as dimers or multimers. Therefore, in certain embodiments, the value of the amide II band, after being calibrated for a particular protein (which may be performed before or during elution), can be used as a trigger to indicate the presence of aggregates, for example, if it deviates from the stable range.

[0487] Figures 29A and 29B show 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 wavenumber region from 1600 to 1700), and the second, amide II peak metric, characterizes the amide II band and is calculated as the area under the curve (AUC) of the amide II band (corresponding approximately to the wavenumber region from 1500 to 1600). The second protein aggregation metric, R, is then calculated as the ratio of the amide I band's area under the curve (AUC) of the amide II band (corresponding approximately to the wavenumber region from 1500 to 1600). Amide-I to Amide-II is calculated as the ratio of AUC of amide I to AUC of amide II (or the ratio of amide II to amide I). Typical ratio values ​​are provided in Tables 3A and 3B below. Highly purified protein monomers typically have ratios in the range of 1.2 to 1.6. Aggregates typically have ratios of 0.8 to 1.0, and fragments have ratios of 0.4 to 0.7. [Table 3A] [Table 3B]

[0488] A third embodiment (Figures 30A and 30B) involves calculating the peak height ratio of amide I to amide II (or alternatively, the ratio of amide II to amide I) where the amide I band approximately corresponds to the 1600-1700 wavenumber region and the amide II band approximately corresponds to the 1500-1600 wavenumber region. Typical ratio values ​​are provided in Figure 26A. Highly purified protein monomers typically have ratios in the range of 1.2-1.6. Aggregates typically have ratios of 0.8-1.0, and fragments have ratios of 0.4-0.7.

[0489] First aggregation metric - ν COM Amide-II was found to provide the highest correlation with aggregation level and performance as a quantitative measure of % aggregation.

[0490] Referring to Figure 31, the total area under the curve of the amide I and amide II bands was used to measure total protein content.

[0491] With reference to Figures 32A and 32B, these sample quality metrics can then be used to generate digital or analog signals that can be used for dynamic process control and / or process optimization. For example, Figure 32B is an exemplary sketch showing the expected (hypothetical) variation in sample quality metrics measuring total protein content and protein aggregation, as described herein, over time during elution from an IEX chromatography column. The solid (black) curve shows the expected variation in the value of the total protein content metric, calculated by integrating over the amide I and amide II band regions, as described in Figure 31. The total protein content is expected to peak when monomeric protein begins to elute from the column. The dashed (black) curve shows protein content as measured by UV absorbance, reflecting variations in mid-IR measurements. The dashed-dotted (red ... COM Amide-II 32B shows the expected variation of a protein aggregation metric, such as v The protein aggregation metric curve initially peaks (3202) as low molecular weight species, such as fragments, are eluted first, then reaches and remains at a relatively stable value over a period of time (3204) as high purity monomer elutes, after which the value shifts to reflect the elution of higher molecular weight species, such as dimers, trimers, etc. (3208). In certain embodiments, as shown in FIG. 32B, v COM Amide-II Use of a protein aggregation metric such as may allow collection over an additional period (3206) during which the total protein content metric, as measured by either IR absorption spectroscopy or UV absorption, begins to decrease, while high purity monomer continues to elute, thereby improving yield. Further discussion of process control based on protein aggregation metrics is provided in Example 10 below. Example 7 Protein Identification and Metric Development Workflow

[0492] This example illustrates an exemplary embodiment of an exemplary workflow that can be used to generate sample quality metrics and identify reference spectra.

[0493] Figures 33A and 33B show examples of methods for creating protein identification methods. Figure 33C provides a schematic 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. Example 8 Monitoring and control of downstream refining processes

[0494] This example provides an example of a control strategy based on the measurement of multiple sample quality metrics to provide real-time control of collection start / stop based on protein purity. Referring to Figure 34A, one exemplary control strategy monitors two sample quality metrics in real time—a total protein content metric and a protein aggregation metric, e.g., total amide band AUC and amide II center of mass as described in Example 6. These two metrics can be used to calculate output analog voltages that trigger the start and stop of collection of desired elution fractions that optimize collected purity (e.g., with respect to monomer).

[0495] In particular, Figure 34A shows the variation of a voltage signal used to control chromatographic elution over time in an IEX column. The first, top trace shows the signal that initiates the conductivity ramp to begin elution. The second trace from the top shows the signal that causes the mid-IR analyzer to record a reference background spectrum, e.g., of water absorbance. As shown in the figure, the background spectrum was recorded just before the conductivity ramp began. After the conductivity ramp began, the figure shows a brief period (3402) (shaded light blue area) where the conductivity increases but before protein begins to elute. This period can be used to calibrate a conductivity-dependent baseline removal function (e.g., as a function of the amount of water displaced for a given conductivity level), if desired. The third trace from the top shows the variation of an analog voltage output that measures total protein content. This voltage output rises and stabilizes as pure monomer begins to elute, and can be used, for example, to trigger the start of collection. Below the total protein content voltage trace is an analog voltage trace proportional to the center of mass frequency of amide II (calculated as described in Example 6 above). As shown in the figure, this voltage is stable while primarily monomeric protein is eluting, then deviates from its tabulated value as dimers begin to elute. Therefore, the voltage of the center of mass of amide II can be used to trigger collection halt.

[0496] 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, "other species" protein metrics can be calculated. In certain embodiments, sample quality metrics measuring 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. Example 9 Monitoring downstream filtration processes

[0497] This example demonstrates the use of the systems and methods described herein to perform accurate real-time quantification of 1-300+ g / L of protein, provide real-time quantification and control of buffer stoichiometry, and provide real-time monitoring of protein aggregation and / or structural changes (e.g., denaturation).

[0498] Figure 35 shows data generated by several sequential injections in a TFF system, including an amide I subband analysis heat map that can be used to visualize secondary structure variation and total protein concentration. Visualization of secondary structure variation in the filtrate and / or retentate from a TFF system can be used to optimize filtration or buffer exchange processes, for example, to determine whether protein aggregation or degradation in secondary structure is occurring. Example 10 Wide spectral range and low noise QCL system

[0499] This example demonstrates the improved performance of a QCL-IR analyzer system according to certain embodiments described herein. For example, Figure 36 shows the IR spectrum at approximately 1025 cm -1 ~Approx. 1725cm -1 Figure 37 shows the use of multiple QCLs to obtain a continuous range of .mu.m. Figure 37 shows the potential for over 10-fold improvement in sensitivity. Example 11 Simultaneous measurement of multiple analytes

[0500] This example demonstrates the simultaneous measurement of multiple analytes in solution with a QCL-based mid-IR spectrometer according to certain embodiments described herein.

[0501] Figures 38A and 38B show measured absorbance spectra of several exemplary analytes in solution, and Figures 39A and 39B show excellent linearity for glucose in water (Figure 39A) and ammonia in water (Figure 39B). Figures 40A and 40B demonstrate the ability to identify and measure target analytes in complex mixtures (e.g., in the presence of other analytes) with a high degree of linearity (RMS concentration error, approximately 10 micrograms / mL). Example 12 Formulation Development and Forced Degradation

[0502] This example demonstrates the use of mid-IR absorption spectroscopy to observe changes in protein secondary structure over time, for example, as relevant to formulation development.

[0503] Figure 41A shows the variation of the protein amide band spectrum with temperature. Without wishing to be bound by any particular theory, it is believed that the band at approximately 1750 cm -1 ~1300cm -1 IR spectra in the range of (e.g., when measured with sufficient sensitivity and / or spectral resolution) provide a wealth of information about intermolecular interactions with side chains and bending of the NH from the amide II peak.

[0504] Figure 41B is a view of the amide I band region of the spectrum shown in Figure 41A, and Figure 41C shows the second derivative spectrum over the amide I band region, which indicates changes in secondary structure, particularly a relative decrease in alpha helix content and an increase in turn content.

[0505] FIG. 42 shows the spectrum of mAb at 10 mg / ml in a particular formulation buffer (blue) along with an analysis of the constituent subbands (red curve).

[0506] Figure 43A shows the variation in amide I band structure for a mAb exhibiting high beta-sheet content as the temperature is increased from 25°C to 80°C. The middle panel is a heat map showing the change in subband intensity with temperature. Figure 43B summarizes the changes in secondary structure motifs and content with increasing temperature, as shown by the superimposed trajectories on the heat map. Example 13 Data automation and stability

[0507] This example demonstrates the automation capabilities and reproducibility and applicability of a mid-IR analyzer as described in certain embodiments herein, suitable for, for example, GMP-compliant bioproduction.

[0508] Figure 44 shows an example of an analytical system including multiple mid-IR analyzers collocated, linked, and connected together with a central computer. In certain embodiments, the mid-IR system is connected to a processor that enables workflow management of multiple instruments and fluid handlers. In certain embodiments, the system provides automated self-diagnostics for wavelength accuracy and light output. In certain embodiments, the system includes a warning system for leaks and / or volatile compounds. In certain embodiments, the system includes an OPC-UA secure communication and instrument control interface that enables compliance with 21 CFR P11.

[0509] Figures 45A-45E show the results of several tests of reproducibility, including high reproducibility, particularly across sequential injections and replicate sets of injections over multiple days. Example 14 Independent protein and nucleic acid quantification and metrics of viral capsid titer and full percentage

[0510] This example demonstrates the use of a QCL-based mid-IR analyzer to independently quantify protein and nucleic acid content and describes several exemplary methods by which protein and nucleic acid content metrics can be used to determine viral capsid concentration and full fraction.

[0511] 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 based on peak intensity metrics, such as peak amplitude or AUC, measured across one or more of the amide I, amide II, and amide III spectral regions. Similarly, the nucleic acid content in a sample can be determined using peak intensity metrics for the asymmetric and / or symmetric PO4 bands. In the context of a viral vector sample containing a protein capsid encapsulating a genetic payload, the protein content measure calculated from the amide I, II, and / or III spectral bands can therefore provide a measure of total capsid content, while the nucleic acid content can be used to confirm the presence of a genetic payload and compared to the capsid content to determine the full capsid fraction.

[0512] Figures 46A and 46B demonstrate the possibility of independent quantification of nucleic acid (in this example, DNA) and protein content in a mixture. Figure 46A plots absorbance spectra measured from pure samples of nucleic acid and protein, specifically ssDNA and bovine serum albumin (BSA). According to various embodiments described herein, the DNA solution spectrum contains absorption peaks in the asymmetric and symmetric P0 bands, as well as features in the amide I region that, without wishing to be bound by any particular theory, result from the presence of C=O groups in the nucleic acid bases (see, e.g., Figure 46C). The spectrum of the BSA solution contains spectral features of amide I and amide II. As shown in Figure 46A, according to certain embodiments described herein, the absorption features of both protein and nucleic acid overlap in the amide I spectral region, while in the amide II region, the spectrum of the BSA solution contains an absorption peak, while the spectrum of the DNA solution is still negligible and relatively flat, demonstrating the usefulness of the amide II peak intensity for quantifying protein content in a mixture. Similarly, for nucleic acid quantification, two strong PO4 (as...

Claims

1. 1. A method for obtaining a purified sample of a target protein species by real-time monitoring of protein heterogeneity and control of the purification process, comprising: (a) measuring, at each of one or more time points, by one or more mid-infrared (MIR) analyzers, a corresponding infrared (IR) absorbance signal from an aqueous sample exiting a purification unit, the aqueous sample containing one or more protein species, including said 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, including a protein aggregation metric indicative of a level of protein aggregation in the aqueous sample; and (d) obtaining the purified sample of the target protein species by using the one or more sample quality metrics to control collection of a target fraction of the aqueous sample. A method comprising:

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

3. 3. The method of 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.

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

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

6. 10. The method of claim 1, wherein the aqueous sample comprises a plurality of different protein species.

7. 10. The method of claim 1, wherein the aqueous sample comprises a heterogeneous population of the target protein species, comprising monomeric and aggregated portions.

8. 10. The method of any one of the preceding claims, wherein the target protein species is a subspecies of a particular protein species, and the target protein species has a particular desired level and / or type of molecular conjugation.

9. At least one of the one or more MIR analyzers comprises: an MIR light source aligned and operable to emit a beam of MIR light; one or more sampling optics that pass through and / or contact at least a portion of the aqueous sample and are aligned to direct and / or allow passage of the beam of MIR light, and / or at least a portion thereof, toward one or more detectors after passing through or contacting the portion of the aqueous sample; and the one or more detectors aligned and operable to detect the beam of MIR light after its passage through and / or contact with the aqueous sample.

10. The method of any one of the preceding claims, which is or comprises an MIR spectrometer comprising:

10. the one or more sampling optics: a high refractive index material aligned such that the beam of MIR light is directed toward, incident on, and internally reflected by an interface between the high refractive index material and the aqueous sample; the one or more detectors are aligned and operable to detect the beam of MIR light exiting therefrom after its internal reflection by the high refractive index material; 10. The method of claim 9.

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

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

13. the one or more sampling optics include a flow cell including a detection channel through which the aqueous sample flows; the one or more detectors are aligned and operable to detect the beam of MIR light exiting therefrom after its transmission through the detection channel; 13. The method according to any one of claims 9 to 12.

14. 14. The method of claim 13, wherein the path length through the detection channel is about 10 μm or longer.

15. 15. The method of any one of claims 9 to 14, wherein the one or more MIR analysis devices is or comprises a quantum cascade laser (QCL)-based MIR spectrometer comprising a QCL light source operable to emit a beam of MIR light.

16. The beam of MIR light is approximately 4 cm －１ 16. The method of any one of claims 9 to 15, wherein the spectral linewidth is equal to or less than 100 .mu.m.

17. 17. The method of any one of claims 9 to 16, wherein the power of the beam of MIR light is about 1 mW or greater.

18. The spectral resolution of the MIR spectrometer is about 4 cm －１ 18. The method of any one of claims 9 to 17, wherein the solubility is 0.01 or better.

19. The accuracy of the MIR spectrometer is about 2 cm －１ 19. The method of any one of claims 9 to 18, wherein the solubility is 0.01 or better.

20. The repeatability of the MIR spectrometer is about 0.5 cm －１ 20. The method of any of claims 9 to 19, wherein the method is or better.

21. and wherein the MIR light source is a tunable laser, and the method comprises, at each of the one or more time points: illuminating the aqueous sample with a plurality of radiation frequencies by sweeping the radiation frequency of the beam MIR light across a scan range of the tunable laser; and detecting the beam of MIR light at each of the plurality of radiation frequencies with the one or more detectors to measure a corresponding infrared (IR) spectrum including a plurality of values ​​as corresponding IR absorbance signals from the aqueous sample, each IR spectrum including a plurality of values ​​associated with, representing, and / or based on the detected output at a particular one of the plurality of radiation frequencies.

21. The method of any one of claims 9 to 20, comprising:

22. The plurality of radiation wavelengths ranges from about 1800 cm to about 800 cm －１ 22. The method of claim 21, comprising one or more wavelengths within a spectral band in the range of

23. The plurality of radiation wavelengths are about 1600 cm －１ ~Approx. 1500cm －１ 23. The method of claim 22, comprising one or more wavelengths within a spectral band in the range of

24. 10. The method of any one of the preceding claims, wherein the MIR analyzer is an online sensor that measures the IR absorbance signal substantially in real time as the aqueous solution exits the purification unit.

25. 2. The method of claim 1, wherein the IR absorbance data comprises a corresponding amide band spectrum for each of the one or more time points.

26. 26. The method of claim 25, wherein step (d) comprises determining, for each particular time point of at least some of said one or more time points, a corresponding value of said protein aggregation metric.

27. determining the corresponding value of the protein aggregation metric for each particular time point, calculating, from the amide band spectrum corresponding to the particular time point, a value of an amide II peak metric that quantifies one or more properties 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.

27. The method of claim 26, comprising:

28. 28. The method of claim 27, wherein the amide II peak metric is a frequency location metric that quantifies the frequency at which the amide II band is substantially centered at the particular time point.

29. 29. The method of claim 28, wherein the frequency location metric is the center of mass frequency of the amide II band.

30. determining the corresponding value of the protein aggregation metric for each particular time point, calculating, from the amide band spectrum corresponding to the particular time point, a value of an amide I peak metric that quantifies one or more characteristics of the amide I band at the particular time point; and determining the corresponding value of the protein aggregation metric using both the amide I peak metric value and the amide II peak metric value.

27. The method of claim 26, comprising:

31. the amide I peak metric is a peak intensity metric that quantifies the intensity of the amide I band at the particular time point; the amide II peak metric is a peak intensity metric that quantifies the intensity of the amide II band at the particular time point; 31. The method of claim 30, wherein determining the value of the protein aggregation metric comprises calculating (i) a ratio of the amide I peak metric value to the amide II peak metric value and / or (ii) a ratio of the amide II peak metric value to the amide I peak metric value.

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

33. 10. The method of any one of the preceding claims, wherein step (d) comprises causing a processor to send one or more trigger signals to a controller unit of the purification unit.

34. 34. The method of 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.

35. 35. The method of 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 a total protein content metric.

36. Step (d) initiating, by the controller unit, collection of the target fraction of the aqueous sample based on the one or more trigger signals; and stopping, by the controller unit, collection of the target fraction of the aqueous sample based on the one or more trigger signals.

36. The method of any one of claims 33 to 35, comprising one or both of:

37. 1. A method for real-time monitoring of protein aggregation in a sample, comprising: (a) repeatedly receiving, by a processor of a computing device, infrared (IR) absorbance data corresponding to IR absorbance signals measured at each of a plurality of time points, the IR absorbance data including, for each particular time point of the plurality of time points, a corresponding IR absorbance spectrum measured from the sample at the particular time point, the IR absorbance data including a plurality of absorbance values ​​each associated with a particular wavenumber; (b) analyzing, by the processor, the IR absorbance data to determine, for each particular time point of the plurality of time points: using the IR absorbance spectrum corresponding to the particular time point to determine values ​​of one or more peak metrics for one or both of an amide I band and an amide II band; using the values ​​of the one or more peak metrics to determine a value of a protein aggregation metric indicative of a level of protein aggregation in the sample at the particular time point; and updating a real-time protein aggregation signal with the determined value of the protein aggregation metric for the particular time point. obtaining the real-time protein aggregation signal, which provides a measure of protein aggregation in the sample as a function of time, by (c) storing and / or providing, by said processor, said real-time protein aggregation signal for one or more of: (i) further processing, (ii) display, and (iii) use as a control signal for regulating one or more purification units. A method comprising:

38. 38. The method of claim 37, wherein step (b) comprises determining, for each particular time point, a value of a frequency position metric that quantifies the frequency at which the amide II band is substantially centered at said particular time point as the value of the protein aggregation metric indicative of the level of protein aggregation in the sample at said particular time point.

39. 39. The method of claim 38, wherein the frequency location metric is the center of mass frequency of the amide II band.

40. Step (b) for each particular time point: determining a value of an amide I peak intensity metric that quantifies the intensity of the amide I band at the particular time point; determining a value of an amide II peak intensity metric that quantifies the intensity of the amide II band at the particular time point; and determining the value of the protein aggregation metric: (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; 40. The method of any one of claims 37 to 39, comprising:

41. 1. A method for mid-IR (MIR) spectroscopy-based monitoring and control of a production unit for the manufacture of a biological product, comprising: (a) measuring, at each of one or more time points, with one or more mid-infrared (MIR) analyzers, a corresponding infrared (IR) absorbance signal from an aqueous sample flowing to and / or from said production unit; (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; 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 production unit. A method comprising:

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

43. 43. The method of 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 flow or normal flow filtration unit, an ultrafiltration unit, and a diafiltration unit.

44. 44. The method of claim 43, wherein the purification unit is a chromatography column.

45. 42. The method of claim 41 , wherein the production unit is a bioreactor.

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

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

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

50. 50. The method of any one of claims 41 to 49, wherein the aqueous sample comprises one or more subspecies of a particular protein species having a particular desired level and / or type of molecular conjugation.

51. 51. The method of any one of claims 41 to 50, wherein the aqueous sample comprises nucleic acids.

52. 52. The method of any one of claims 41 to 51, wherein the aqueous sample comprises one or more species of virus and / or virus-like particles.

53. 53. The method of any one of claims 41 to 52, wherein step (c) comprises using the IR absorbance data to determine values ​​of one or more sample quality metrics at each of the one or more time points.

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

55. 55. The method of claim 53 or 54, wherein the one or more sample quality metrics comprises a protein aggregation metric indicative of a level of protein aggregation within the aqueous sample.

56. 56. The method of any one of claims 53 to 55, wherein the one or more sample quality metrics comprise one or more protein species metrics that identify the presence of and / or quantify the content of one or more specific protein species within the aqueous sample.

57. 57. The method of any one of claims 53 to 56, wherein the one or more sample quality metrics comprises a protein conjugation metric that quantifies the level and / or type of molecular conjugation.

58. 58. The method of any one of claims 53 to 57, wherein the one or more sample quality metrics comprise one or more protein secondary structure metrics that quantify the presence and / or content of one or more protein secondary structure motifs.

59. 59. The method of any one of Claims 53 to 58, wherein the one or more sample quality metrics comprise one or more nucleic acid content metrics that quantify the content of nucleic acids within the aqueous sample.

60. 60. The method of any one of claims 53 to 59, wherein the one or more sample quality metrics comprise one or more virus content metrics that quantify the content of viruses and / or virus-like particles within the aqueous sample.

61. 61. The method of any one of claims 53 to 60, wherein the one or more sample quality metrics comprise one or more empty / full metrics that quantify the content and / or relative proportion of empty and / or full viral vectors within the aqueous sample.

62. 62. The method of any one of claims 53-61, wherein the one or more sample quality metrics comprises a capsid aggregation metric indicating a level of capsid aggregation within the viral vector sample.

63. 63. The method of any one of claims 53-62, wherein the one or more sample quality metrics comprises a viral nucleic acid content metric that distinguishes the viral nucleic acid from the host cell protein and host cell nucleic acid content.

64. 63. The method of any one of claims 53 to 62, wherein at least a portion 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.

65. The IR absorbance data is (i) one or more amide II absorbance values ​​that relate to and measure IR absorption at wavenumbers within the amide II spectral band; and / or (ii) one or more amide III absorbance values ​​that correlate to and measure IR absorption at wavenumbers within the amide II spectral band; 65. The method of any one of claims 53 to 64, comprising:

66. 66. The method of claim 65, wherein determining the one or more sample quality metrics comprises determining a value of a protein content metric that quantifies protein content within the sample based at least in part on the amide II and / or amide III absorbance values.

67. determining a value of an amide II peak metric based on the amide II absorbance value and / or a value of an amide III peak metric based on the amide III absorbance value; and determining the protein content metric using the amide II peak metric and / or the amide III peak metric.

67. The method of claim 66, comprising:

68. 68. The method of claim 67, wherein the amide II peak metric and / or the amide III peak metric are peak intensity metrics that quantify the intensity of the amide II band and / or the amide III band, respectively.

69. The IR absorbance data is (i) Asymmetric PO ４ One or more asymmetric phosphate stretches (asymmetric POs) are associated with and measured by IR absorption at wavenumbers within a spectral band. ４ ) absorbance value; and / or (ii) Symmetric PO ４ One or more symmetric phosphate stretches (symmetric POs) associated with and measured by IR absorption at wavenumbers within a spectral band. ４ ) absorbance value 69. The method of any one of claims 53 to 68, comprising:

70. The step of determining the one or more sample quality metrics may further comprise: ４ and / or symmetric PO ４ 70. The method of claim 69, comprising determining a value of a nucleic acid content metric that quantifies nucleic acid content in the sample based at least in part on the absorbance value.

71. The asymmetric PO ４ Asymmetric PO based on absorbance values ４ The value of the peak metric and / or the symmetry PO ４ Symmetric PO based on absorbance values ４ determining a value of a peak metric; and The asymmetric PO ４ Peak metric value and / or the symmetric PO ４ using the peak metric to determine the nucleic acid content metric.

71. The method of claim 70, comprising:

72. The asymmetric PO ４ Peak metric and / or the symmetric PO ４ The peak metrics are respectively ４ Band and / or symmetric PO ４ 72. The method of claim 71, wherein the peak intensity metric quantifies the intensity of the band.

73. 73. The method of any one of Claims 53-72, wherein determining the one or more sample quality metric values ​​comprises independently quantifying total protein and nucleic acid content within the sample by determining (i) a value of a protein content metric that quantifies protein content within the sample and (ii) a value of a nucleic acid (e.g., DNA, RNA, etc.) content metric that quantifies nucleic acid content within the sample.

74. 74. The method of Claim 73, wherein determining the one or more sample quality metric values ​​comprises determining a total capsid content based at least in part on the value of the protein content metric.

75. 75. The method of Claim 73 or 74, wherein determining the one or more sample quality metric values ​​comprises determining a full capsid fraction 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.

76. 76. The method of any one of claims 53 to 75, wherein the IR absorbance data is or comprises one or more IR absorbance spectra, each IR absorbance spectrum comprising, for each particular wavenumber of a plurality of wavenumbers across a measured spectral band, a corresponding IR absorbance value representing a measure of absorption of IR light by the aqueous sample at the particular wavenumber.

77. The measured spectral bands are an amide II band, an amide III band, an asymmetric PO ４ bands, and symmetric PO ４ 77. The method of claim 76, wherein the bands are across one or more bands selected from the group consisting of:

78. 78. The method of any one of claims 53-77, comprising monitoring the one or more sample quality metrics over time by determining values ​​of the one or more sample quality metrics for each of the one or more time points.

79. 80. The method of claim 78, comprising determining values ​​of the one or more sample quality metrics in substantially real time.

80. 80. The method of any one of claims 53 to 79, wherein at least one particular sample quality metric of the one or more sample quality metrics is calculated using a machine learning model that receives one or more IR spectra as input and generates the particular sample quality metric as output.

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

82. the IR absorbance data comprises an IR absorbance spectrum, and step (c) comprises: receiving one or more reference spectra, each measured from a corresponding reference sample; and determining the values ​​of at least some of the one or more sample quality metrics using the IR absorbance spectrum and the one or more reference spectra; 82. The method of any one of claims 53 to 81, comprising:

83. 83. The method of claim 82, wherein the one or more reference spectra comprise high quality viral vector spectra measured from a reference sample having a full capsid fraction at or above a particular threshold fraction.

84. 84. The method of claim 83, wherein the threshold percentage is about 75%.

85. 85. The method of 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 absorbance spectrum.

86. 86. The method of 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 absorbance spectrum.

87. Step (c) a correlation value based on a correlation between (i) a particular one of said one or more reference spectra and / or one or more derivatives thereof and (ii) said IR absorbance spectrum and / or one or more derivatives thereof; a covariance value based on the covariance between (i) a particular one of said one or more reference spectra and / or one or more derivatives thereof and (ii) said IR absorbance spectrum and / or one or more derivatives thereof; (i) a Pearson correlation value between a particular one of said one or more reference spectra and / or one or more derivatives thereof and (ii) said IR absorbance spectrum and / or one or more derivatives thereof; 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 derivatives thereof and (ii) the IR absorbance spectrum and / or one or more derivatives thereof; 87. The method of any one of claims 82 to 86, comprising calculating one or more members selected from the group consisting of:

88. Step (c) obtaining a set of sample peak metric values ​​by determining values ​​for a set of one or more specific peak metrics from the IR absorbance spectrum; and determining one or more values ​​of the sample quality metrics based on the set of sample peak metric values ​​and a set of reference peak metric values ​​determined for the one or more particular peak metrics from the one or more reference spectra.

88. The method of any one of claims 82 to 87, comprising:

89. 89. The method of any one of claims 82 to 88, comprising determining a similarity score measuring the similarity between the one or more reference spectra and the IR absorbance spectrum.

90. 90. The method of any one of claims 82 to 89, comprising monitoring deviations from the one or more reference spectra in real time by repeatedly performing steps (a) to (d) in substantially real time.

91. 91. The method of any one of claims 53 to 90, wherein the one or more time points is a plurality of time points.

92. 92. The method of 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 sample quality metric using values ​​of the first sample quality metric corresponding to two or more of the plurality of time points.

93. 93. The method of claim 92, wherein the second sample quality metric is a time difference metric that measures the change in the first sample quality metric over time and is calculated based on the difference between (i) values ​​of the first sample quality metric at a first set of time points and (ii) values ​​of the first sample quality metric at a second set of time points.

94. 94. The method of claim 92 or 93, wherein the second sample quality metric is a time-aggregated signal that is at least in part a function of the plurality of time points.

95. 95. The method of any one of claims 41 to 94, wherein step (c) comprises causing the processor to send one or more trigger signals to a controller unit of the production unit.

96. 96. The method of claim 95, wherein the one or more trigger signals comprise an analog voltage signal having a time-varying amplitude based at least in part on the value of one or more sample quality metrics.

97. 97. The method of any one of claims 41 to 96, wherein step (c) comprises using a machine learning model for adjusting the one or more process parameters.

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

99. 99. The method of any one of claims 41 to 98, wherein the one or more process parameters include amounts of one or more raw materials.

100. 100. The method of any one of claims 41 to 99, wherein the one or more process parameters include times to start and / or stop sub-processes.

101. the one or more IR absorbance signals to which the IR absorbance data received in step (b) correspond are measured from the aqueous sample at each of one or more time points as the aqueous sample exits the purification unit; The method comprises: using the IR absorption data to determine one or both of (i) total capsid content and (ii) percent full capsid; and obtaining a purified sample of viral vector material by controlling collection of a target fraction of said aqueous sample using said determined total capsid content and / or full capsid percentage.

101. The method of any one of claims 41 to 100, comprising:

102. the one or more IR absorbance signals to which the IR absorbance data received in step (b) correspond are measured from the aqueous sample at each of one or more time points as the aqueous sample exits the purification unit; The method comprises: 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 obtaining a purified sample of viral vector material by controlling collection of a target fraction of said aqueous sample using said capsid aggregation metric.

102. The method of any one of claims 41 to 101, comprising:

103. 1. A system for obtaining purified samples of target protein species by real-time monitoring of protein heterogeneity and control of purification processing, comprising: (a) one or more mid-infrared (MIR) analyzers aligned and operable to measure, at each of one or more time points, a corresponding infrared (IR) absorbance signal from an aqueous sample exiting the purification unit, the aqueous sample containing one or more protein species, including said target protein species; (b) the processor of the computing device; and (c) Memory in which instructions are stored the instructions, when executed by the processor, cause the processor to: receiving IR absorbance data corresponding to the IR absorbance signal at each of the one or more time points; determining values ​​of one or more sample quality metrics based on the IR absorbance spectrum, including one or more sample quality metrics, wherein the one or more sample quality metrics include a protein aggregation metric indicative of a level of protein aggregation in the aqueous sample; providing and / or using the one or more sample quality metrics to control collection of a target fraction of the aqueous sample, thereby obtaining the purified sample of the target protein species.

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

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

106. 1. A system for real-time monitoring of protein aggregation in a sample, comprising: the processor of the computing device; and Memory where instructions are stored the instructions, when executed by the processor, cause the processor to: (a) repeatedly receiving infrared (IR) absorbance data corresponding to IR absorbance signals measured at each of a plurality of time points, the IR absorbance data including, for each particular time point of the plurality of time points, a corresponding IR absorbance spectrum measured from the sample at the particular time point, the IR absorbance data including a plurality of absorbance values ​​each associated with a particular wavenumber; (b) analyzing the IR absorbance data to, for each particular time point of the plurality of time points: using the IR absorbance spectrum corresponding to the particular time point to determine values ​​of one or more peak metrics for one or both of an amide I band and an amide II band; using the values ​​of the one or more peak metrics to determine a value of a protein aggregation metric indicative of a level of protein aggregation in the sample at the particular time point; and updating a real-time protein aggregation signal with the determined value of the protein aggregation metric for the particular time point. obtaining said real-time protein aggregation signal, which provides a measure of protein aggregation in said sample as a function of time, by (c) storing and / or providing said real-time protein aggregation signal for one or more of: (i) further processing, (ii) display, and (iii) use as a control signal to regulate one or more purification units; system.

107. 1. A system for monitoring and control based on mid-IR (MIR) spectroscopy of a production unit for the manufacture of biological products, comprising: (a) one or more mid-infrared (MIR) analyzers aligned and operable to measure, at each of one or more time points, a corresponding infrared (IR) absorbance signal from an aqueous sample flowing to and / or from said production unit; (b) the processor of the computing device; and (c) a memory having instructions stored therein that, when executed by the processor, cause the processor to: receiving IR absorbance data corresponding to the IR absorbance signal at each of the one or more time points; a memory for causing the received IR absorbance data to be used to adjust one or more process parameters of the production unit; Including, the system.

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

109. 1. A method for quantifying and / or monitoring viral vector quality in an aqueous sample containing one or more species of virus and / or virus-like particles, comprising: (a) receiving, by a processor of a computing device, infrared (IR) absorbance data corresponding to one or more IR absorbance signals measured from the sample; (b) determining, by the processor, the value of one or more viral vector sample quality metrics using the IR absorbance data; and (c) storing and / or providing said one or more viral vector sample quality metrics for display and / or further processing. A method comprising:

110. 110. The method of claim 109, wherein the one or more viral vector quality metrics comprises a total capsid content, which quantifies the content of viral capsids within the sample.

111. 111. The method of claim 109 or claim 110, wherein the one or more viral vector sample quality metrics comprises full capsid percentage.

112. 112. The method of any one of claims 109-111, wherein the one or more viral vector sample quality metrics comprises a capsid aggregation metric indicating a level of capsid aggregation within the viral vector sample.

113. 113. The method of any one of claims 109-112, wherein the one or more viral vector sample quality metrics comprises a protein content metric that quantifies protein content within the sample.

114. 114. The method of any one of claims 109-113, wherein the one or more viral vector sample quality metrics comprises a nucleic acid content metric that quantifies nucleic acid content within the sample.

115. 115. The method of any one of claims 109-114, wherein the one or more viral vector sample quality metrics comprises a viral nucleic acid content metric that distinguishes the viral nucleic acid from the host cell protein and host cell nucleic acid content.

116. Step (b) determining, by the processor, a value for each of one or more peak metrics of the IR absorption data, each peak metric being associated with one or more particular spectral bands and quantifying a particular structural characteristic of one or more absorption peaks within the particular spectral bands; and using the determined values ​​of the one or more peak metrics to determine the values ​​of at least some of the viral vector sample quality metrics.

116. The method of any one of claims 109 to 115, comprising:

117. The IR absorbance data is (i) one or more amide II absorbance values ​​that relate to and measure IR absorption at wavenumbers within the amide II spectral band; and / or (ii) one or more amide III absorbance values ​​that correlate to and measure IR absorption at wavenumbers within the amide II spectral band; 117. The method of any one of claims 109 to 116, comprising:

118. 118. The method of claim 117, wherein step (b) comprises determining a value of a protein content metric that quantifies protein content in the sample based at least in part on the amide II and / or amide III absorbance values.

119. determining a value of an amide II peak metric based on the amide II absorbance value and / or a value of an amide III peak metric based on the amide III absorbance value; and determining the protein content metric using the amide II peak metric and / or the amide III peak metric.

119. The method of claim 118, comprising:

120. 120. The method of claim 119, wherein the amide II peak metric and / or the amide III peak metric are peak intensity metrics that quantify the intensity of the amide II band and / or the amide III band, respectively.

121. 121. The method of any one of claims 109-120, wherein the one or more viral vector sample quality metrics comprise one or more protein structural metrics indicative of the presence and / or abundance of one or more particular protein structural forms within the sample.

122. The IR absorbance data is (i) Asymmetric PO ４ One or more asymmetric phosphate stretches (asymmetric POs) are associated with and measured by IR absorption at wavenumbers within a spectral band. ４ ) absorbance value; and / or (ii) Symmetric PO ４ One or more symmetric phosphate stretches (symmetric POs) associated with and measured by IR absorption at wavenumbers within a spectral band. ４ ) absorbance value 122. The method of any one of claims 109 to 121, comprising:

123. Step (b) is ４ and / or symmetric PO ４ 123. The method of Claim 122, comprising determining a value of a nucleic acid content metric that quantifies nucleic acid content in the sample based at least in part on the absorbance value.

124. The asymmetric PO ４ Asymmetric PO based on absorbance values ４ The value of the peak metric and / or the symmetry PO ４ Symmetric PO based on absorbance values ４ determining a value of a peak metric; and The asymmetric PO ４ Peak metric value and / or the symmetric PO ４ using the peak metric to determine the nucleic acid content metric.

124. The method of claim 123, comprising:

125. The asymmetric PO ４ Peak metric and / or the symmetric PO ４ The peak metrics are respectively ４ Band and / or symmetric PO ４ 125. The method of claim 124, wherein the peak intensity metric quantifies the intensity of the band.

126. 126. The method of any one of Claims 109-125, wherein step (b) comprises independently quantifying total protein and nucleic acid content in the sample by determining (i) a value of a protein content metric that quantifies protein content in the sample and (ii) a value of a nucleic acid content metric that quantifies nucleic acid content in the sample.

127. 127. The method of Claim 126, wherein step (b) comprises determining, as one of the viral vector sample quality metrics, total capsid content based at least in part on the value of the protein content metric.

128. 128. The method of Claim 126 or 127, wherein step (b) comprises determining, as one of the viral vector sample quality metrics, a full capsid fraction 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.

129. 129. The method of any one of claims 109 to 128, wherein the IR absorbance data is or includes one or more IR absorbance spectra, each IR absorbance spectrum including, for each particular wavenumber of a plurality of wavenumbers across a measured spectral band, a corresponding IR absorbance value representing a measure of absorption of IR light by the aqueous sample at the particular wavenumber.

130. The measured spectral bands are an amide II band, an amide III band, an asymmetric PO ４ bands, and symmetric PO ４ 130. The method of claim 129, spanning one or more bands selected from the group consisting of bands.

131. step (a) comprising repeatedly receiving the IR absorbance data at a plurality of time points to obtain, for each of the plurality of time points, a corresponding set of IR absorbance data; 131. The method of any one of claims 109 to 130, wherein the method comprises monitoring the total capsid content and / or full capsid fraction over time by performing steps (b)-(c) for each set of IR absorbance data.

132. 132. The method of claim 131, comprising performing steps (a) through (c) in substantially real time to obtain: (i) a real time capsid content signal that provides a measure of capsid content in the sample as a function of time, and / or a fraction full capsid signal that provides a measure of the fraction of capsids in the sample that are full, as a function of time.

133. 133. The method of any one of claims 109 to 132, comprising measuring the one or more IR absorbance signals by one or more mid-infrared (MIR) analyzers.

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

135. 135. A method according to any one of claims 133 to 134, comprising measuring the one or more IR absorbance signals derived from the aqueous sample as it flows to and / or from a production unit.

136. 136. The method of any one of claims 109 to 135, wherein the one or more species of virus and / or virus-like particle comprises one or more species of adeno-associated virus (AAV).

137. 137. The method of any one of claims 109 to 136, wherein the one or more species of virus include adenovirus and / or retrovirus.

138. 138. The method of any one of claims 109 to 137, wherein the one or more species of virus comprises a plant-based virus.

139. 139. A method according to any one of claims 109 to 138, wherein the one or more IR absorbance signals to which the IR absorbance data corresponds are measured from the aqueous sample as it flows to and / or from a production unit.

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

141. 141. The method of 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 flow or normal flow filtration unit, an ultrafiltration unit, and a diafiltration unit.

142. 142. The method of claim 141, wherein the purification unit is a chromatography column.

143. 143. The method of any one of claims 139 to 142, wherein the production unit is or comprises a bioreactor.

144. 144. The method of any one of claims 109 to 143, wherein step (c) comprises causing the processor to generate and / or transmit one or more trigger signals to a control unit of the production unit based at least in part on one or more of the determined viral vector sample quality metrics.

145. 145. The method of claim 144, wherein step (c) comprises causing the processor to generate a trigger signal having a value based at least in part on the determined viral vector quality metric.

146. the one or more IR absorbance signals to which the IR absorbance data received in step (a) corresponds are measured from the aqueous sample at each of one or more time points as the aqueous sample exits a purification unit; The method comprises: using the IR absorption data to determine one or both of (i) total capsid content and (ii) percent full capsid; and obtaining a purified sample of viral vector material by controlling collection of a target fraction of said aqueous sample using said determined total capsid content and / or full capsid percentage.

146. The method of any one of claims 109 to 145, comprising:

147. the one or more IR absorbance signals to which the IR absorbance data received in step (a) corresponds are measured from the aqueous sample at each of one or more time points as the aqueous sample exits a purification unit; The method comprises: 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 obtaining a purified sample of viral vector material by controlling collection of a target fraction of said aqueous sample using said capsid aggregation metric.

147. The method of any one of claims 109 to 146, comprising:

148. the IR absorbance data comprises an IR absorbance spectrum, and step (b) comprises: receiving one or more reference spectra, each measured from a corresponding reference sample comprising the target viral vector species and / or one or more model components thereof at a high purity and / or concentration; and determining the values ​​of at least some of the one or more viral vector sample quality metrics using the IR absorbance spectrum and the one or more reference spectra; 148. The method of any one of claims 109 to 147, comprising:

149. 149. The method of claim 148, wherein the one or more reference spectra comprise high-quality viral vector spectra measured from a reference sample having a full capsid fraction at or above a particular threshold fraction.

150. 150. The method of claim 149, wherein the threshold percentage is approximately 75%.

151. 151. The method of any one of claims 148 to 150, wherein step (b) comprises calculating a difference spectrum based on at least one of the one or more reference spectra and the IR absorbance spectrum.

152. 152. A method according to any one of claims 148 to 151, wherein step (b) comprises calculating one or more derivative spectra of at least one of the one or more reference spectra and / or the IR absorbance spectrum.

153. Step (b) a correlation value based on a correlation between (i) a particular one of said one or more reference spectra and / or one or more derivatives thereof and (ii) said IR absorbance spectrum and / or one or more derivatives thereof; a covariance value based on the covariance between (i) a particular one of said one or more reference spectra and / or one or more derivatives thereof and (ii) said IR absorbance spectrum and / or one or more derivatives thereof; (i) a Pearson correlation value between a particular one of said one or more reference spectra and / or one or more derivatives thereof and (ii) said IR absorbance spectrum and / or one or more derivatives thereof; 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 derivatives thereof and (ii) the IR absorbance spectrum and / or one or more derivatives thereof; calculating one or more members selected from the group consisting of 153. The method of any one of claims 148 to 152, comprising:

154. Step (b) obtaining a set of sample peak metric values ​​by determining values ​​for a set of one or more specific peak metrics from the IR absorbance spectrum; and determining said measure of deviation based on said set of sample peak metric values ​​and a set of reference peak metric values ​​determined for said one or more particular peak metrics from said one or more reference spectra.

154. The method of any one of claims 148 to 153, comprising:

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

156. 156. A method according to any one of claims 148 to 155, comprising the step of monitoring deviations from the one or more reference spectra in real time by repeatedly performing steps (a) to (c) in substantially real time.

157. 1. A method for assessing and / or monitoring the quality of viral vector content in an aqueous sample containing a target viral vector species, comprising: (a) receiving, by a processor of a computing device, infrared (IR) absorbance data corresponding to one or more IR absorbance signals measured from the sample, the IR absorbance data comprising an IR absorbance spectrum measured from the sample and including a plurality of absorbance values ​​each associated with a particular wavenumber; (b) receiving, by said processor, one or more reference spectra, each measured from a corresponding reference sample comprising said target viral vector species and / or one or more model components thereof at a high purity and / or concentration; (c) determining, by the processor, one or more measures of deviation using the IR absorbance spectrum and the one or more reference spectra; (d) storing and / or providing said measure of deviation for display and / or further processing. A method comprising:

158. 1. A system for quantifying and / or monitoring viral vector quality in an aqueous sample containing one or more species of virus and / or virus-like particles, comprising: the processor of the computing device; and A memory having stored thereon instructions which, when executed by the processor, cause the processor to perform the method of any one of claims 109 to 157. Including, the system.

159. 1. A method for monitoring compositional changes of a sample by infrared (IR) absorption spectroscopy, comprising: (a) repeatedly receiving, by a processor of a computing device, infrared (IR) absorbance data corresponding to IR absorbance signals measured at each of a plurality of time points, the IR absorbance data including, for each particular time point of the plurality of time points, a corresponding IR absorbance spectrum measured from the sample at the particular time point, the IR absorbance data including a plurality of absorbance values ​​each associated with a particular wavenumber; (b) analyzing, by the processor, the IR absorbance data to determine, for each particular time point of the plurality of time points: obtaining a current normalized spectrum by normalizing the current IR absorbance spectrum corresponding to the particular time point using reference absorbance values ​​determined from values ​​of the current IR absorbance spectrum at one or more reference wavenumbers; determining a current value of the spectral difference metric based on a difference between the current normalized spectrum and a prior normalized spectrum, the prior normalized spectrum being based on one or more previously acquired IR absorption spectra each corresponding to and measured at a particular prior time point, each particular previously acquired IR absorption spectrum having been normalized using a reference value determined from values ​​of the particular previously acquired IR absorbance spectrum at the one or more reference wavenumbers; and updating a real-time normalized spectral difference signal according to the current value of the normalized spectral difference metric; obtaining a normalized spectral difference signal that measures the change in normalized spectral absorbance between successive time points by: (c) storing and / or providing, by said processor, said real-time normalized spectral difference signal for one or more of: (i) further processing, (ii) display, and (iii) use as a control signal for adjusting one or more process parameters of a production unit. A method comprising:

160. 160. The method of claim 159, further comprising identifying changes in the composition of the sample based on the real-time normalized spectral difference signal.

161. 161. The method of claim 160, comprising detecting change points in the real-time normalized spectral difference signal and identifying the change in the composition of the sample based on the detected change points.

162. 162. The method of claim 160 or 161, comprising determining values ​​of one or more statistical parameters of the real-time normalized spectral difference signal.

163. The one or more statistical parameters are: average; Dispersion; Mode; and Standard deviation 163. The method of claim 162, comprising one or more members selected from the group consisting of:

164. 164. The method of claim 162 or 163, comprising identifying the change in composition based on the value of at least one of the one or more statistical parameters (i) exceeding one or more thresholds and / or (ii) changing outside a specified range.

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

166. 10. The method of claim 1, wherein the sample comprises one or more protein species.

167. 167. The method of any one of claims 166, further comprising identifying a change in the composition of the sample corresponding to a change in the purity and / or properties of the target protein species.

168. 168. The method of claim 167, wherein the identified change in composition is or comprises a change in the level and / or presence of protein aggregates within the aqueous sample.

169. The identified change in composition is a change in the content of one or more particular protein species within said sample; Changes in the level and / or type of molecular conjugation; and Alteration of the content of one or more protein secondary structure motifs 169. The method of claim 167 or 168, wherein said method is or comprises one or more members selected from the group consisting of:

170. 10. The method of claim 1, wherein the sample comprises nucleic acids.

171. 171. The method of Claim 170, further comprising identifying a change in the composition of said sample corresponding to a change in the purity and / or properties of said nucleic acid within said sample.

172. 10. The method of any one of the preceding claims, wherein the aqueous sample comprises one or more species of virus and / or virus-like particles.

173. 173. The method of claim 172, further comprising identifying a change in the composition of said sample corresponding to a change in the purity and / or characteristics of said virus and / or virus-like particle within said sample.

174. 174. The method of claim 173, wherein the change in composition corresponds to a change in the relative proportion of empty and / or full viral vectors within the aqueous sample.

175. The method of claim 173 or claim 174, wherein the change in composition corresponds to the level of capsid aggregation within the sample.

176. 176. The method of any one of claims 173 to 175, wherein said change in composition corresponds to a change in the relative content between viral nucleic acid derived from host cell proteins and host cell nucleic acid content.

177. 177. A method according to any one of claims 160 to 176, comprising causing adjustment of one or more process parameters of a production unit based on said identification of said change in the composition of said sample.

178. 178. The method of claim 177, wherein the production unit is or includes a purification unit.

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

180. 180. The method of any one of claims 177 to 179, wherein the production unit is or includes one or more members selected from the group consisting of a flow controller, a valve controller, and a temperature controller profile.

181. 181. A method according to any one of claims 177 to 180, comprising the step of triggering a response of a production unit.

182. 182. The method of claims 178 to 181, wherein the sample is an aqueous sample and the method comprises a step of obtaining a purified sample by causing adjustment of a collection window to control collection of a target fraction of the aqueous sample.

183. 183. The method of any one of claims 178 to 182, wherein the production unit is or includes a filtration unit.

184. 184. The method of any one of claims 177 to 183, comprising a step of monitoring the progress of a chemical reaction based on the real-time normalized spectral difference signal.

185. 185. The method of any one of claims 177 to 184, comprising a step of causing adjustment of one or more members selected from the group consisting of in-line buffer preparation, a mixing process, and a temperature controller.

186. 10. The method of claim 9, wherein the reference absorbance value is determined from a value of the current IR absorbance spectrum at a single reference wavenumber, and the previous reference value is determined from a value of the previous IR absorbance spectrum at the single reference wavenumber.

187. 10. The method of claim 9, wherein determining the current value of the spectral difference metric comprises calculating an integrated absorbance over one or more specific spectral bands for each of the current normalized spectrum and the previous normalized spectrum.

188. the one or more particular spectral bands are Amide I spectral band, and / or from about 1500 to about 1600 cm －１ Amide II region in the range of Amide II spectral band; and Amide III spectral bands 188. The method of claim 187, comprising one or more members selected from the group consisting of:

189. the one or more particular spectral bands are Asymmetric PO ４ spectral band; and Symmetric PO ４ Spectral bands 189. The method of claim 187 or claim 188, comprising one or more members selected from the group consisting of:

190. 10. The method of claim 1, further comprising measuring the IR absorbance signal using one or more MIR analyzers.

191. 1. A method for monitoring temporal changes in a sample by infrared (IR) absorption spectroscopy, comprising: (a) repeatedly receiving, by a processor of a computing device, infrared (IR) absorbance data corresponding to IR absorbance signals measured at each of a plurality of time points, the IR absorbance data including, for each particular time point of the plurality of time points, a corresponding IR absorbance spectrum measured from the sample at the particular time point, the IR absorbance data including a plurality of absorbance values ​​each associated with a particular wavenumber; (b) analyzing, by said processor, said IR absorbance data; a time difference signal measuring the temporal change between values ​​of one or more properties determined using (i) a first set of one or more IR absorbance spectra corresponding to a first set of specific time points and (ii) a second set of one or more IR absorbance spectra corresponding to the first set of specific time points; and a time-aggregated signal that is a function at least in part of the plurality of time points; obtaining one or both of: (c) storing and / or providing, by said processor, said time difference signal and / or said time aggregated signal for one or more of: (i) further processing, (ii) display, and (iii) use as a control signal for adjusting one or more process parameters of a production unit. A method comprising:

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

193. 193. The system of claim 192, further comprising one or more MIR analysis devices.

194. 194. The system of claim 192 or 193, including a production unit.

195. 1. A method for obtaining purified samples of target protein species by mid-infrared (IR) spectroscopy-based bioprocess monitoring and control, comprising: (a) measuring a plurality of mid-infrared (MIR) absorbance spectra over time by one or more mid-infrared analyzers, by measuring, at each of a plurality of time points, a corresponding mid-IR absorbance spectrum from an aqueous sample exiting a purification unit, the aqueous sample containing one or more protein species, including the target protein species; (b) receiving, by a processor of a computing device, spectral data corresponding to the plurality of measured mid-IR absorbance spectra; (c) determining, by the processor, for each of at least some of the plurality of time points, based on the spectral data, a corresponding value of one or more sample quality metrics comprising a measure of concentration and / or purity of the target protein species in the aqueous sample; and (d) obtaining the purified sample of the target protein species by using the determined values ​​of the one or more sample quality metrics to control collection of a target fraction of the aqueous sample. A method comprising:

196. 196. The method of claim 195, wherein the purification unit is or comprises a chromatography column.

197. 197. The method of claim 195 or 196, wherein the purification unit is or comprises an ultrafiltration and diafiltration system (UF / DF).

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

199. 200. The method of any one of claims 195 to 198, wherein the aqueous sample comprises a plurality of different molecular forms of a specific protein, including a monomeric form and one or more aggregated forms, and the target protein species is the monomeric form of the specific protein.

200. 200. The method of any one of claims 195 to 199, wherein the aqueous sample comprises one or more subspecies of a particular protein, each having a particular desired level and / or type of molecular conjugation, and the target protein species is a particular one of the one or more subspecies.

201. the one or more MIR analyzers: a QCL-based light source aligned and operable to emit a beam of MIR light; one or more sampling optics that pass through and / or contact at least a portion of the aqueous sample and are aligned to direct and / or allow passage of the beam of MIR light, and / or at least a portion thereof, toward one or more detectors after passing through or contacting the portion of the aqueous sample; and the one or more detectors aligned and operable to detect the beam of MIR light after its passage through and / or contact with the aqueous sample.

201. The method of any one of claims 195 to 200, comprising a mid-IR spectrometer based on a quantum cascade laser (QCL) comprising:

202. the one or more sampling optics include a flow cell including a detection channel through which the aqueous sample flows; the one or more detectors are aligned and operable to detect the beam of MIR light exiting therefrom after its transmission through the detection channel; The method of claim 201.

203. and wherein the QCL-based light source is a tunable QCL that operatively sweeps the emission frequency of the beam of MIR light through a plurality of frequencies over a scan range, and the method comprises, at each of the one or more time points: illuminating the aqueous sample with a plurality of radiation frequencies by sweeping the radiation frequency of the beam MIR light across the scan range of the tunable laser; and detecting the beam of MIR light at each of the plurality of radiation frequencies with the one or more detectors to measure a corresponding infrared (IR) spectrum including a plurality of values ​​as corresponding IR absorbance signals from the aqueous sample, each IR spectrum including a plurality of values ​​associated with, representing, and / or based on the detected output at a particular one of the plurality of radiation frequencies.

203. The method of claim 201 or 202, comprising:

204. A method according to any one of claims 195 to 203, wherein the MIR analysis device is an online sensor and step (a) comprises repeatedly measuring the IR absorbance spectrum over time (e.g., every 20 seconds or less) as the aqueous solution exits the purification unit.

205. 205. The method of any one of claims 195 to 204, wherein the spectral data comprises a corresponding amide band spectrum for each of the one or more time points.

206. Step (c) receiving, by the processor, a reference spectrum for the target protein species, the reference measured from a specific corresponding reference sample comprising the target protein species in a substantially isolated and / or highly purified form; and tracking the concentration of the target protein species over time by repeatedly using the reference spectrum 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.

206. The method of any one of claims 195 to 205, comprising:

207. Step (c) receiving, by the processor, reference spectra of one or more impurities, each associated with a particular impurity of interest and measured from a particular corresponding reference sample containing the impurity of interest in a substantially isolated and / or highly purified form; and repeatedly using the reference spectra of the one or more impurities at each of the plurality of time points to determine the concentration of each impurity of interest in the aqueous sample.

207. The method of any one of claims 195 to 206, comprising:

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

209. 209. The method of any one of claims 195 to 208, wherein step (d) comprises causing the processor to send one or more trigger signals to a controller unit of the purification unit and / or to downstream (from the purification unit) valves.

210. Step (d) initiating, by the controller unit, collection of the target fraction of the aqueous sample based on the one or more trigger signals; and stopping, by the controller unit, collection of the target fraction of the aqueous sample based on the one or more trigger signals.

210. The method of claim 209, comprising one or both of:

211. wherein the target protein species is a monomeric form of a particular protein, and the method comprises: In step (c), determining (i) the concentration of the monomeric form of the specific protein and / or (ii) the cumulative purity value of the monomeric form of the specific protein over time within the total collected volume of the sample exiting the purification unit; and In step (d), stopping collection of the aqueous sample exiting the purification unit at a specified stop time based at least in part on the concentration and / or the value of cumulative purity of the monomeric form of the specified protein.

211. The method of any one of claims 195 to 210, comprising:

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

213. the aqueous sample comprises one or more additives, and the method further comprises: In step (c), determining concentration and / or amount values ​​of the one or more additives in the aqueous sample exiting the purification unit at one or more time points based on the spectral data; and In step (d), using the determined value of the concentration and / or amount of additive to control collection of the target fraction of the aqueous sample.

213. The method of any one of claims 195 to 212, comprising:

214. 1. A method for preparing a biologic drug formulation containing one or more additives, comprising: (a) receiving a solution containing a purified drug substance that includes a protein species; (b) injecting and / or mixing one or more additives into said solution of said purified drug substance over a period of time to create an in-process drug substance solution comprising said purified drug substance and said one or more additives in relative concentrations that vary over said period of time as said one or more additives are injected and / or mixed; (c) measuring, at each of one or more time points, by one or more mid-infrared (MIR) analyzers: (i) the corresponding mid-IR absorbance spectrum from said in-process drug substance solution; and (ii) the corresponding mid-IR absorbance spectrum from a stock solution containing at least one of said one or more additives. measuring one or more mid-IR absorbance spectra by measuring one or both of: (d) receiving, by a processor of a computing device, spectral data corresponding to the one or more measured mid-IR absorbance spectra; (e) determining, by the processor, for each of at least some of the one or more time points based on the spectral data, corresponding values ​​of one or more sample quality metrics, the one or more sample quality metrics comprising a measure of concentration and / or purity of (i) the protein species and / or (ii) a subset of the one or more additives; and (f) using the determined values ​​of the one or more sample quality metrics to control the dosing and / or mixing of the one or more additives to obtain a final drug substance having a desired protein and / or additive content and / or purity. A method comprising:

215. 215. The method of claim 214, wherein step (b) comprises using an ultrafiltration / diafiltration (UF / DF) system.

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

217. 217. The method of any one of claims 214 to 216, wherein the one or more additives are or include one or more surfactants {e.g., detergents; e.g., wetting agents and / or solubilizing agents}.

218. 218. The method of any one of claims 214 to 217, wherein the one or more additives are or include one or more bulking agents {e.g., sugars and / or polyols; e.g., amino acids; e.g., polymers and proteins}.

219. the one or more MIR analyzers: a QCL-based light source aligned and operable to emit a beam of MIR light; one or more sampling optics passing through and / or contacting at least a portion of the stock solution and / or a portion of the in-process drug substance solution and aligned to direct and / or allow passage of the beam of MIR light, and / or at least a portion thereof, towards one or more detectors after passing through or contacting the portion of the stock solution and / or the portion of the in-process drug substance solution; and the one or more detectors aligned and operable to detect the beam of MIR light after its passage through and / or contact with the portion of the stock solution and / or the portion of the in-process drug substance solution.

219. The method of any one of claims 214 to 218, comprising a mid-IR spectrometer based on a quantum cascade laser (QCL), comprising:

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

221. and wherein the QCL-based light source is a tunable QCL operable to sweep the emission frequency of the beam of MIR light through a plurality of frequencies over a scan range, and the method comprises, at each of the one or more time points: illuminating the portion of the stock solution and / or the portion of the in-process drug substance solution with multiple radiation frequencies by sweeping the radiation frequency of the beam of MIR light across the scan range of the tunable laser; and detecting the beam of MIR light at each of the plurality of radiation frequencies with the one or more detectors to measure a corresponding infrared (IR) spectrum comprising a plurality of values ​​as the corresponding IR absorbance signal from the portion of the stock solution and / or the portion of the in-process drug substance solution, each IR spectrum being associated with, representative of, and / or based on the detected output at a particular one of the plurality of radiation frequencies.

221. The method of claim 219 or 220, comprising:

222. 222. A method according to any one of claims 214 to 221, wherein the MIR analysis device is an online sensor and step (c) comprises repeatedly measuring the IR absorbance spectrum over time (e.g., every 20 seconds or less) as one or more additives are injected and / or mixed.

223. 223. The method of any one of claims 214 to 222, wherein the spectral data comprises a corresponding amide band spectrum for each of the one or more time points.

224. 224. The method of any one of claims 214 to 223, wherein the spectral data comprises a corresponding sugar band spectrum for each of the one or more time points.

225. Step (e) receiving, by the processor, a reference spectrum for the protein species, the reference measured from a specific corresponding reference sample comprising the substantially isolated and / or highly purified protein species; and tracking the concentration of the protein species over time by repeatedly using the reference spectrum 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.

225. The method of any one of claims 214 to 224, comprising:

226. Step (e) receiving, by the processor, one or more additive reference spectra, each associated with a particular additive of interest and measured from a particular corresponding reference sample containing the particular additive of interest in a substantially isolated and / or highly purified form; and repeatedly using the one or more additive reference spectra at each of the plurality of time points to determine the concentration of each of the one or more additives of interest in the stock solution and / or in-process drug substance solution.

226. The method of any one of claims 214 to 225, comprising:

227. 227. The method of any one of claims 214 to 226, wherein step (f) comprises causing the processor to send one or more trigger signals to a controller unit.

228. Step (f) initiating, by the controller unit, the injection and / or mixing of the one or more additives; and stopping, by the controller unit, the injection and / or mixing of the one or more additives based on the one or more trigger signals.

228. The method of claim 227, comprising one or both of:

229. 1. A system for obtaining purified samples of target protein species by mid-infrared (IR) spectroscopy-based bioprocess monitoring and control, comprising: one or more mid-infrared (MIR) analyzers; the processor of the computing device; and Memory where instructions are stored the instructions, when executed by the processor, cause the processor to: (a) receiving spectral data corresponding to a plurality of measured mid-IR absorbance spectra from an aqueous sample exiting a purification unit, the aqueous sample including one or more protein species including the target protein species, the mid-IR absorbance spectra being each measured by the one or more MIR analyzers at a corresponding one of a plurality of time points; (b) determining, for each of at least some of the plurality of time points, a corresponding value of one or more sample quality metrics based on the spectral data, the value comprising a measure of concentration and / or purity of the target protein species in the aqueous sample; (c) obtaining the purified sample of the target protein species by using the determined values ​​of the one or more sample quality metrics to control collection of a target fraction of the aqueous sample. system.

230. 1. A system for preparing a biologic drug formulation containing one or more excipients, comprising: one or more mid-infrared (MIR) analyzers; the processor of the computing device; and Memory where instructions are stored the instructions, when executed by the processor, cause the processor to: (a) at each of one or more time points, by the one or more MIR analyzers: (i) an in-process drug substance solution comprising a purified drug substance and one or more additives that are injected and / or mixed therein / with over time; and (ii) a stock solution containing at least one of said one or more additives; receiving spectral data corresponding to one or more mid-IR absorbance spectra measured from one or both of the (b) determining, for each of at least some of the one or more time points, corresponding values ​​of one or more sample quality metrics based on the spectral data, the one or more sample quality metrics comprising a measure of concentration and / or purity of (i) the protein species and / or (ii) a subset of the one or more additives; (c) using the determined values ​​of the one or more sample quality metrics to control the injection and / or mixing of the one or more additives to obtain a final drug substance having a desired protein and / or additive content and / or purity. system.

Images