Systems and methods for ion population regulation in mass spectrometry

Abstract

A method of performing mass spectrometry includes accumulating ions resulting from components eluted from a chromatographic column over an accumulation time and transferring the accumulated ions to a mass analyzer. During an acquisition, a mass spectrum is acquired of detected ions originating from the transferred ions. An elution profile is obtained from a series of acquired mass spectra including the acquired mass spectrum and a plurality of previously acquired mass spectra. The elution profile includes a plurality of detection points representing a change in intensity of the detected ions over time. A current signal state of the elution profile is classified based on a subset of detection points included in the plurality of detection points. The accumulation time is set for a next acquisition of a mass spectrum based on the classified current signal state of the elution profile.

Description

[0001] Cross Reference to Related Applications

[0002] This application claims priority to U.S. Patent Application No. 17 / 135,749, filed December 28, 2020, the contents of which are hereby incorporated by reference in their entirety. Background Technology

[0003] A mass spectrometer is a sensitive instrument used to detect, identify, and / or quantify molecules based on their mass-to-charge ratio (m / z). A mass spectrometer typically comprises an ion source for generating ions from components contained in a sample, a mass analyzer for separating ions based on their m / z, and an ion detector for detecting the separated ions. The mass spectrometer can be connected to a computer-based software platform that uses data from the ion detector to construct a mass spectrum showing the relative abundance of each ion among the detected ions as a function of m / z. The m / z of the ions can be used to detect and quantify molecules in simple and complex mixtures. Separation devices such as liquid chromatography (LC) or gas chromatography (GC) can be coupled to a mass spectrometer (MS) in a combined system (e.g., an LC-MS or GC-MS system) to separate components contained in a sample before introducing them into the mass spectrometer.

[0004] For trap-type mass analyzers, such as quadrupole ion traps and orbitraps TM Electrostatic trapping mass analyzers (manufactured by Thermo Fisher Scientific of Waltham, MA) typically achieve optimal analytical performance when the number of ions being analyzed is within a target range. Too few ions can lead to spectral noise or require the addition of multiple spectra, while too many ions can cause space charge effects such as mass shift, peak broadening, and coalescence. Beam-type instruments, such as time-of-flight (ToF) and quadrupole mass filters, also operate optimally when the ion flux at their detector is within a certain range. Both trapping and beam-type devices can be used with intermediate ion storage traps to accumulate ions and buffer downstream processes such as mass analysis, thereby improving scan speed and instrument sensitivity. These ion storage traps also have limited capacity, and overfilling can lead to detrimental effects such as mass discrimination and loss of linearity, as the amount of detected signal is no longer linearly related to the ion accumulation time.

[0005] Automatic gain control (AGC) can be performed to regulate the ion population in trapping and beam-type devices. When the trapping device is filled with a target number of ions, conventional AGC can be performed using a gate device that transmits or blocks ions. Using an estimate of the ion flux, the gate is opened for a given time to accumulate the target number of ions in the device, and then the gate is closed until the accumulated ions are transferred out of the device. Alternatively, and particularly in the case of beam-type devices, attenuation devices or partial gates can be used to regulate the ion flux to a level that does not saturate the analyzer or detector.

[0006] In traditional AGC methods, the accumulation time used to acquire mass spectra is estimated based on the ion flux of the most recently acquired mass. However, traditional AGC methods assume that the ion flux will remain constant in the next acquisition, and therefore cannot account for the rapid changes in ion flux (e.g., ion intensity) during LC-MS or GC-MS analysis. Consequently, traditional AGC methods may exceed the target ion number when the ion flux increases, and may fall below the target ion number when the ion flux decreases, both of which can degrade analytical performance. The drawbacks of traditional AGC techniques are exacerbated when the ion flux exhibits a non-Gaussian curve, where the ion flux rapidly increases from the baseline level to a peak value and then gradually decreases back to the baseline level. In these cases, the number of ions accumulated during the increase in ion flux may be significantly greater than the number of ions accumulated when the ion flux exhibits a Gaussian curve.

[0007] One approach to addressing these issues is to use a conventional AGC (Automatic Guided Collection) scheme while reducing the time between analytical scans. This brute-force method offers some improvement in ion population control accuracy, but at the cost of efficiency. For example, when an AGC scheme uses a dedicated pre-scan to estimate ion flux, the time spent on the pre-scan cannot be used to acquire the analytical scan. When an AGC scheme uses a previous analytical scan to estimate ion flux, the analytical scan must be repeated more frequently, thereby reducing instrument capacity. Furthermore, increasing the sampling rate is not entirely effective unless the analysis is limited by a small number of analytes.

[0008] For at least these reasons, improved methods and systems are needed to modulate ion populations in mass spectrometry. Summary of the Invention

[0009] The following description presents a simplified overview of one or more aspects of the methods and systems described herein in order to provide a basic understanding of such aspects. This overview is not a comprehensive review of all aspects covered, and is not intended to identify key or decisive elements of all aspects, nor to define the scope of any or all aspects. Its sole purpose is to present some concepts of one or more aspects of the methods and systems described herein in a simplified form as an introduction to the more detailed descriptions presented below.

[0010] In some illustrative embodiments, a method of performing mass spectrometry includes: accumulating ions generated from components eluted from a chromatographic column over a cumulative time period; transferring the accumulated ions to a mass analyzer; acquiring a mass spectrum of detected ions originating from the transferred ions during an acquisition period; obtaining an elution curve from a series of acquired mass spectra including the acquired mass spectrum and a plurality of previously acquired mass spectra, the elution curve including a plurality of detection points representing changes in the intensity of the detected ions over time; classifying the current signal state of the elution curve based on a subset of detection points included in the plurality of detection points; and setting the cumulative time for the next acquisition of the mass spectrum based on the classified current signal state of the elution curve.

[0011] In some illustrative embodiments, obtaining the mass spectrum includes performing a pre-scan.

[0012] In some illustrative embodiments, obtaining the mass spectrum includes performing tandem mass spectrometry; and the detected ions include product ions.

[0013] In some illustrative embodiments, the elution curves are based on the total ion current of each mass spectrum contained in the series of mass spectra.

[0014] In some illustrative embodiments, classifying the current signal state of the elution curve includes selecting a specific signal state from a plurality of signal states.

[0015] In some illustrative embodiments, each signal state indicates how the intensity of the detected ion changes over time.

[0016] In some illustrative embodiments, the plurality of signal states include a baseline state, a rising state, a peak state, and a falling state.

[0017] In some illustrative embodiments, setting the cumulative time based on the classified current signal state of the elution curve includes: determining a nominal cumulative time; identifying a correction factor associated with the classified current signal state of the elution curve; and correcting the nominal cumulative time based on the correction factor associated with the classified current signal state of the elution curve.

[0018] In some illustrative embodiments, the nominal cumulative time is determined based on the target ion population divided by the current ion flux.

[0019] In some illustrative embodiments, the correction factor is configured to optimize the returns given by the reward strategy.

[0020] In some illustrative embodiments, the correction factor is determined at least in part based on data obtained from the series of acquired mass spectra.

[0021] In some illustrative embodiments, the current signal state of the elution curve is classified based on the relative position of selected detection points included in the plurality of detection points, the relative position of which indicates the position of the selected detection point relative to a desired reference point in the elution curve.

[0022] In some illustrative embodiments, the selected detection point is the latest detection point.

[0023] In some illustrative embodiments, the relative position of the selected detection point includes the normalized intensity value of the selected detection point, which represents the ratio of the intensity value detected at the selected detection point to the expected maximum intensity value of the elution curve.

[0024] In some illustrative embodiments, the relative position of the selected detection point includes the time distance of the selected detection point from the expected time point in the elution curve.

[0025] In some illustrative embodiments, classifying the current signal state of the elution curve includes applying the subset of detection points to a trained machine learning model configured to estimate the relative positions of the selected detection points based on the subset of detection points.

[0026] In some illustrative embodiments, the machine learning model is trained at least in part based on the plurality of detection points of the elution curve.

[0027] In some illustrative embodiments, an apparatus for performing mass spectrometry includes: an ion accumulator configured to accumulate ions generated from components eluted from a chromatographic column over an accumulation time; a mass analyzer configured to acquire a series of mass spectra of detected ions originating from the accumulated ions during an acquisition period after the accumulated ions have been transferred from the ion accumulator; and a computing device configured to: obtain an elution curve from the series of mass spectra including the acquired mass spectra and a plurality of previously acquired mass spectra, the elution curve including a plurality of detection points representing changes in the intensity of the detected ions over time; classify the current signal state of the elution curve based on a subset of detection points included in the plurality of detection points; and set the accumulation time for the next acquisition of the mass spectra based on the classified current signal state of the elution curve.

[0028] In some illustrative embodiments, a method includes: obtaining an elution curve from a series of mass spectra including a latest mass spectrum and a plurality of previously acquired mass spectra, the elution curve including a plurality of detection points, each detection point representing the change in intensity over time of ions accumulated and detected by a detector within a cumulative time; classifying the current signal state of the elution curve based on a subset of detection points included in the plurality of detection points; and setting the cumulative time for the next acquisition of the mass spectrum based on the classified current signal state of the elution curve.

[0029] In some illustrative embodiments, a method of performing mass spectrometry includes: acquiring a mass spectrum of ions accumulated over a cumulative time; and setting the cumulative time for the next acquisition of a mass spectrum based on a series of mass spectra including the mass spectrum and a plurality of previously acquired mass spectra. Attached Figure Description

[0030] The accompanying drawings illustrate various embodiments and are part of the specification. The illustrated embodiments are merely examples and do not limit the scope of this disclosure. Throughout the drawings, the same or similar reference numerals denote the same or similar elements.

[0031] Figure 1 An illustrative LC-MS system comprising a liquid chromatograph and a mass spectrometer is shown.

[0032] Figure 2A It shows a device containing an ion accumulator. Figure 1 An illustrative implementation scheme for a mass spectrometer.

[0033] Figure 2B This diagram shows a capture-type quality analyzer. Figure 1 Another illustrative implementation of the mass spectrometer.

[0034] Figure 3 It shows when the component is from Figure 1 Illustrative elution curves (e.g., total ion current (TIC) chromatograms) of ions generated by components contained in a sample during elution in a liquid chromatograph.

[0035] Figure 4 An illustrative mass spectrometry control system configured to perform ion population regulation is shown.

[0036] Figure 5 An illustrative method for performing ion population regulation is shown.

[0037] Figure 6 An illustrative TIC including elution curves is shown, which can be obtained from... Figure 1 The LC-MS system obtains a series of mass spectra that generate or represent the series of mass spectra and contain multiple detection points.

[0038] Figure 7 This shows what happens after two additional detection points are obtained. Figure 6 The illustrative TIC.

[0039] Figure 8 An illustrative method for estimating the relative position of a selected detection point with respect to a reference point in the elution curve is shown.

[0040] Figure 9 This shows the normalized intensity value based on the reference intensity point in the elution curve. Figure 7 The elution curve is divided into several different signal state regions for illustration.

[0041] Figure 10 This displays a time reference point based on time and relative to the elution curve, which will be... Figure 7 The elution curves of TIC shown are another illustrative division into different signal state regions.

[0042] Figure 11 An illustrative method for training a machine learning model that can implement a signal state model is shown.

[0043] Figure 12 It shows that it can be used for training Figure 11 The training data for the machine learning model is generated or the descriptive TIC representing the training data is presented.

[0044] Figure 13 An illustrative method is shown for setting the cumulative time based on the current signal state of the ion signal from the elution curve.

[0045] Figure 14Another illustrative method is shown for setting the accumulation time based on the current signal state of the ion signal from the elution curve.

[0046] Figure 15 An illustrative reward strategy is shown that can be used to determine the correction factor associated with multiple different signal states.

[0047] Figure 16 This shows the correction factor C that depicts the total return determined by the reward strategy as a function of multiple segments. rising A table illustrating the changes.

[0048] Figure 17 An illustrative computing device is shown. Detailed Implementation

[0049] This document describes methods and systems for ion population control in mass spectrometry. In some illustrative embodiments, the method for performing ion population control includes accumulating ions generated from components eluted from a column over a cumulative time and transferring the accumulated ions to a mass analyzer. During acquisition, a mass spectrum originating from the transferred ions is acquired and detected by a detector. From a series of acquired mass spectra including the acquired mass spectra and multiple previously acquired mass spectra, an elution curve comprising multiple detection points, each representing a change in the intensity of a detected ion over time, is obtained. Based on a subset of the detection points included in the multiple detection points, the current signal state of the elution curve is classified, and a cumulative time is set based on the current signal state of the classified elution curve for the next mass spectrum acquisition.

[0050] Various embodiments will now be described in more detail with reference to the drawings. The systems and methods described herein may provide one or more of the benefits described above and / or various additional and / or alternative benefits will become apparent herein.

[0051] In some embodiments, methods and systems for performing ion population regulation can be used in conjunction with combined separation-mass spectrometry systems, such as LC-MS systems. Therefore, exemplary LC-MS systems will now be described. The described LC-MS systems are illustrative and not limiting. The methods and systems described herein can be operated as part of or in conjunction with the LC-MS systems described herein and / or with any other suitable separation-mass spectrometry systems, including high-performance liquid chromatography-mass spectrometry (HPLC-MS), gas chromatography-mass spectrometry (GC-MS), or capillary electrophoresis-mass spectrometry (CE-MS) systems. The methods and systems described herein can also be operated with any other continuous flow sample source, such as a flow injection MS system (FI-MS), in which the analyte is injected into the mobile phase (without separation in the column) and enters the mass spectrometer in a manner that changes intensity over time (e.g., a Gaussian peak).

[0052] Figure 1 An illustrative LC-MS system 100 is shown. The LC-MS system 100 includes a liquid chromatograph 102, a mass spectrometer 104, and a controller 106. The liquid chromatograph 102 is configured to separate components (e.g., analytes) injected into a sample 108 over time. The sample 108 may contain, for example, chemical components (e.g., molecules, ions, etc.) and / or biological components (e.g., metabolites, proteins, lipids, etc.) for detection and analysis by the LC-MS system 100. The liquid chromatograph 102 can be implemented by, for example, any liquid chromatograph conforming to a particular embodiment. In the liquid chromatograph 102, the sample 108 may be injected into a mobile phase (e.g., a solvent) that carries the sample 108 through a column 110 containing a stationary phase (e.g., an adsorbent packing material). As the mobile phase passes through column 110, components within sample 108 are eluted from column 110 at different times based on, for example, their size, their affinity for the stationary phase, their polarity and / or their hydrophobicity.

[0053] A detector (e.g., an ion detector assembly, ion-electron converter, and electron multiplier of mass spectrometer 104) can measure the relative intensity of the signal modulated by each separated component in the eluent 112 from column 110. The data generated by the detector can be represented as a chromatogram plotting retention times on the x-axis and signals representing relative intensity on the y-axis. The retention time of a component is typically measured as the time interval between the injection of sample 108 into the mobile phase and the appearance of the peak of maximum relative intensity after chromatographic separation. In some instances, the relative intensity can be correlated with or represent the relative abundance of the separated component. The data generated by liquid chromatograph 102 can be output to controller 106.

[0054] In some cases, specifically in the analysis of complex mixtures, multiple distinct components in sample 108 may co-elute from column 110 at approximately the same time and therefore may have similar retention times. Therefore, determining the relative intensities of the individual components within sample 108 requires further separation of the individual components. For this purpose, liquid chromatograph 102 directs the components contained in eluent 112 to mass spectrometer 104 for identification and / or quantification of one or more components.

[0055] Mass spectrometer 104 is configured to generate ions from components received by liquid chromatograph 102 and to classify or separate the generated ions based on their m / z. A detector in mass spectrometer 104 measures the intensity of the signal generated by the ions. As used herein, “intensity” or “signal intensity” can refer to any suitable measure, such as abundance, relative abundance, ion count, intensity, or relative intensity. The data generated by the detector can be represented by a mass spectrum, which plots the intensity of the observed signal as a function of the m / z of the detected ions. Data acquired by mass spectrometer 104 can be output to controller 106.

[0056] Mass spectrometer 104 can be implemented by any suitable mass spectrometer. Figure 2A A functional diagram of an illustrative embodiment 200A of a mass spectrometer 104 is shown. As illustrated, the mass spectrometer 104 includes an ion source 202, an ion accumulator 204, a mass analyzer 206, a detector 208, and a controller 210. The mass spectrometer 104 may further include any additional or alternative components, not shown, that may be suitable for a particular embodiment (e.g., ion optics, filters, ion storage devices, ion mobility analyzers, etc.).

[0057] Ion source 202 is configured to generate an ion stream 212 from the components eluted from liquid chromatograph 102 and deliver the ion stream 212 to ion accumulator 204. Ion source 202 can use any suitable ionization technique, including but not limited to electron ionization, chemical ionization, matrix-assisted laser desorption / ionization, electrospray ionization, atmospheric pressure chemical ionization, atmospheric pressure photoionization, inductively coupled plasma, etc. Ion source 202 may include various components for generating ions from the components contained in sample 108 and delivering the ions to ion accumulator 204.

[0058] Ion accumulator 204 is a device configured to accumulate ions contained in ion stream 212 over an accumulation time. In some instances, ion accumulator 204 is configured to buffer downstream processes, such as ion storage devices for mass analysis, thereby improving scan speed and instrument sensitivity. In some instances, ion accumulator 204 is a trapping device, such as a linear quadrupole ion trap, a three-dimensional quadrupole ion trap, a cylindrical ion trap, a toroidal ion trap, an orbital electrostatic trap, a Kingdon trap, etc. The accumulation of ions in ion accumulator 204 can be adjusted to achieve a target number of ions in ion accumulator 204. The adjustment of ion accumulation in ion accumulator 204 can be performed by a gate device (not shown) that transmits or blocks ion stream 212. Using an estimate of the ion flux of ion stream 212 (e.g., the number of ions per unit time (e.g., by intensity measurement)), as will be described in more detail below, the gate can be opened for a given amount of time to dose an appropriate number of ions, and then the gate can be closed. The accumulated ions can then be transferred from the ion accumulator 204 to the mass analyzer 206 as an ion stream 214.

[0059] Mass analyzer 206 is configured to separate ions in ion stream 214 according to the m / z of each ion to filter and / or perform ion mass analysis and frequently provides ion stream 216 to detector 208. Mass analyzer 206 can be implemented by any suitable beam-type or trap-type mass analyzer, such as quadrupole mass filters, ion traps (e.g., linear quadrupole ion traps, three-dimensional quadrupole ion traps, cylindrical ion traps, toroidal ion traps, etc.), time-of-flight (TOF) mass analyzers, electrostatic trap mass analyzers (e.g., orbital electrostatic traps, such as orbitrap mass analyzers, Kinton traps, etc.), Fourier transform ion cyclotron resonance (FT-ICR) mass analyzers, sector mass analyzers, etc.

[0060] In some instances, mass spectrometer 104 is a tandem mass spectrometer (time-tandem or spatial-tandem) configured to perform tandem mass spectrometry (e.g., MS / MS), or a multistage mass spectrometer (also referred to as MS). nThis can be a multistage mass spectrometer or a mixed mass spectrometer. For example, mass analyzer 206 may include multiple mass analyzers, mass filters, and / or collision chambers. As used herein, the term "collision chamber" can encompass any structure arranged to generate product ions via a controlled dissociation process, and is not limited to means for collision-activated dissociation. For example, a collision chamber may be configured to fragment ions using collision-induced dissociation (CID), electron transfer dissociation (ETD), electron capture dissociation (ECD), photoinduced dissociation (PID), surface-induced dissociation (SID), etc. The collision chamber may be positioned upstream of a mass filter that separates fragmented ions based on their mass-to-charge ratio. In some embodiments, mass analyzer 206 may include a combination of multiple mass filters and / or collision chambers, such as a triple quadrupole mass analyzer, wherein the collision chamber is inserted into the ion path between independently operable mass filters.

[0061] and Figure 2A Ion accumulator 204 is shown located upstream of mass analyzer 206. Ion accumulator 204 can be located at any other location within a tandem mass spectrometer or multistage mass spectrometer along the ion path from ion source 202 to detector 208 (e.g., between the first mass filter (Q1) and collision chamber (Q2) and / or between collision chamber (Q2) and second mass filter (Q3)). Alternatively, mass spectrometer 104 may contain more than one ion accumulator 204, as when mass spectrometer 104 is a tandem mass spectrometer or multistage mass spectrometer.

[0062] Ion detector 208 is configured to detect ions at various m / z in or from the ion stream 216 of mass analyzer 206 and responsively generate an electrical signal representing the ion intensity. The electrical signal is transmitted to controller 210 for processing, such as constructing a mass spectrum of the detected ions. For example, mass analyzer 206 may emit an emission beam of separated ions to detector 208, which is configured to detect ions in the emission beam and generate or provide data that can be used by controller 210 to construct a mass spectrum. Ion detector 208 can be implemented using any suitable detection device, including but not limited to electron multipliers, Faraday cups, etc.

[0063] Controller 210 may be communicatively coupled to mass spectrometer 104 and configured to control the operation of said mass spectrometer. For example, controller 210 may be configured to control the operation of various hardware components included in ion source 202, ion accumulator 204, mass analyzer 206, and / or detector 208. For illustration, controller 210 may be configured to control the accumulation time of ion accumulator 204 and / or mass analyzer 206, control the oscillating voltage power supply and / or DC power supply to supply RF voltage and / or DC voltage to mass analyzer 206, adjust the values ​​of RF voltage and DC voltage to select the effective m / z (including the mass tolerance window) for analysis, and adjust the sensitivity of ion detector 208 (e.g., by adjusting detector gain).

[0064] The controller 210 may also include and / or provide a user interface configured to enable interaction between the user of the mass spectrometer 104 and the controller 210. The user may interact with the controller 210 via the user interface through tactile, visual, auditory, and / or other sensory communication. For example, the user interface may include a display device (e.g., a liquid crystal display (LCD) screen, a touchscreen, etc.) for displaying information (e.g., mass spectra, notifications, etc.) to the user. The user interface may also include an input device (e.g., a keyboard, mouse, touchscreen device, etc.) allowing the user to provide input to the controller 210. In other instances, the display device and / or input device may be discretely but communicatively coupled to the controller 210. For example, the display device and input device may be contained in a computer (e.g., a desktop computer, laptop computer, mobile device, etc.) communicatively connected to the controller 210 via a wired connection (e.g., via one or more cables) and / or a wireless connection (e.g., Wi-Fi, Bluetooth, near-field communication, etc.).

[0065] Controller 210 may include any suitable hardware (e.g., processor, circuitry, etc.) and / or software that can serve a particular implementation. Although Figure 2A The controller 210 is shown to be included within the mass spectrometer 104; however, the controller 210 may alternatively be implemented entirely or partially separate from the mass spectrometer 104, such as via a computing device communicatively coupled to the mass spectrometer 104 via a wired connection (e.g., cable) and / or a network (e.g., local area network, wireless network (e.g., Wi-Fi), wide area network, Internet, cellular data network, etc.). In some instances, the controller 210 may be implemented entirely or partially via the controller 106 (and vice versa).

[0066] Figure 2BA functional diagram of another illustrative embodiment 200B of the mass spectrometer 104 is shown. Embodiment 200B is similar to embodiment 200A, except that in embodiment 200B, the mass analyzer 206 is a capture-type mass analyzer and the ion accumulator 204 and the ion stream 214 from the ion accumulator 204 are omitted. Ions accumulate in the mass analyzer 206 during the accumulation time to reach the target ion quantity in the mass analyzer 206.

[0067] Refer again Figure 1 The controller 106 may be communicatively coupled to the LC-MS system 100 (e.g., liquid chromatograph 102 and / or mass spectrometer 104) and configured to control the operation of the LC-MS system. The controller 106 may comprise any suitable hardware (e.g., processor, circuitry, etc.) and / or software configured to control the operation of various components of the LC-MS system 100 (e.g., liquid chromatograph 102 and / or mass spectrometer 104) and / or interact with said various components.

[0068] For example, controller 106 can be configured to acquire data over time by liquid chromatograph 102 and mass spectrometer 104. The data can comprise a series of mass spectra containing the intensity values ​​of ions produced by components of sample 108, varying with the m / z of the ions. The data can be represented in a three-dimensional plot where time (e.g., retention time) is plotted along the x-axis, m / z along the y-axis, and intensity along the z-axis. Spectral features (e.g., peaks of intensity) on the plot represent the detection of ions produced by the various components contained in sample 108 by the LC-MS system 100. The x-axis and z-axis of the plot can be used to generate a mass chromatogram that plots the intensity of a selected m / z (e.g., extracted ion chromatogram (XIC)) or a full m / z spectrum (e.g., total ion current (TIC)) over time. As used herein, “selected m / z” can include a specific m / z with or without a mass tolerance window or a narrower range of m / z. The y-axis and z-axis of the graph can be used to generate mass spectra, plotting the intensity of each mass spectrum as a function of m / z for a specific acquisition (e.g., for each MS scan or MS / MS scan).

[0069] Figure 3A portion of an illustrative mass chromatogram 300 (e.g., TIC) of ions derived from components contained in sample 108 and eluted from liquid chromatograph 102 is shown. Mass chromatogram 300 is generated from data acquired by mass spectrometer 104, such as multiple survey acquisitions (e.g., MS full-spectrum pre-scan) or analytical acquisitions (e.g., MS or MS / MS scan). Mass chromatogram 300 plots the intensity (in arbitrary units) as a function of retention time (in minutes). As shown, mass chromatogram 300 contains multiple detection points, each acquired from a different acquisition, which together form an elution curve 302 for the component (indicated by the dashed curve). When the component elutes from column 110, the intensity of the detected ions produces a peak 304 with a generally Gaussian curve. However, peak 304 and / or other peaks (not shown) in elution curve 302 may have other non-Gaussian curves.

[0070] Traditional AGC methods used to regulate ion populations in ion storage devices or mass analyzers assume that the ion flux (the number of ions per unit time (e.g., intensity)) will remain constant at the next acquisition, and therefore the accumulation time is set based on the previously acquired ion flux. However, as Figure 3 As shown, the ion flux during LC-MS experiments is not constant but dynamic, and sometimes changes rapidly as components are eluted from the column. Therefore, conventional AGC methods exceed the accumulation time during ion accumulation (and thus, the target ion number) when peak intensity increases, and do not exceed the accumulation time (and thus, the target ion number) when peak intensity decreases.

[0071] An improved method for regulating the ion population explains the dynamic and rapidly changing nature of ion flux by classifying the current signal state of the ion signal in the elution curve based on a set of historical detection points (e.g., the most recent set of detection points), and sets the accumulation time for the next acquisition based on the current signal state of the ion signal in the elution curve. For example, the nominal accumulation time t for the next acquisition. nom The target number of ions to be accumulated can be determined by dividing the previously acquired ion flux by the number of ions to be accumulated. The nominal accumulation time t can then be corrected using a correction factor associated with the current signal state of the ion signal in the elution curve. nom To set the cumulative time t for the next acquisition accum For example, when the current signal state is rising (e.g., the ion intensity is increasing), compared to the nominal cumulative time t... nom In comparison, the correction factor reduces the cumulative time t. accum To prevent exceeding the target ion count. When the current signal state is in a declining state (e.g., ion strength is decreasing), the nominal cumulative time t is used. nom In comparison, the correction factor can increase the cumulative time t.accum This is to prevent the number of ions from falling below the target ion count. These and other illustrative methods and systems for regulating the ion population will be described in more detail below.

[0072] One or more operations associated with the improved method of ion population control can be performed by a mass spectrometry control system. Figure 4 The illustrated mass spectrometry control system 400 (“System 400”) is shown. System 400 may be implemented wholly or partially by LC-MS system 100 (e.g., by controller 106 and / or controller 210). Alternatively, System 400 may be implemented separately from LC-MS system 100.

[0073] System 400 may include, but is not limited to, storage facility 402 and processing facility 404 selectively and communicatively coupled to each other. Facilities 402 and 404 may each include or be implemented by hardware and / or software components (e.g., processor, memory, communication interface, instructions stored in memory for execution by the processor, etc.). In some instances, facilities 402 and 404 may be distributed among multiple devices and / or multiple locations that may serve an implementation scheme.

[0074] Storage facility 402 may maintain (e.g., store) executable data that processing facility 404 uses to perform any of the operations described herein. For example, storage facility 402 may store instructions 406 that can be executed by processing facility 404 to perform any of the operations described herein. Instructions 406 may be implemented by any suitable application, software, code, and / or other instance of executable data.

[0075] Storage facility 402 may also maintain any data acquired, received, generated, managed, used, and / or transmitted by processing facility 404. For example, storage facility 402 may maintain LC-MS data (e.g., acquired chromatogram data and / or mass spectrometry data) and / or classification data. Classification data may include data representing one or more models (e.g., machine learning models) or algorithms maintained by processing facility 404 for classifying the current signal state of ion signals in elution curves, or data used by said one or more models or algorithms, or data associated with said one or more models or algorithms.

[0076] Processing facility 404 can be configured to perform (e.g., execute instructions 406 stored in storage facility 402 to perform) the various processing operations described herein. It will be appreciated that the operations and examples described herein are merely illustrative of the many different types of operations that can be performed by processing facility 404. In the following description, any reference to operations performed by system 400 should be understood as being performed by processing facility 404 of system 400. Furthermore, in the description herein, any operation performed by system 400 should be understood to include system 400 instructing or directing another system or apparatus to perform the operation.

[0077] Figure 5 An illustrative method for ion population control in mass spectrometry is shown. Although Figure 5 The illustration shows illustrative operation according to one embodiment, but other embodiments may omit, add, reorder, and / or modify it. Figure 5 Any of the operations shown. Figure 5 One or more of the operations shown can be performed by LC-MS system 100 and / or system 400, any components contained therein and / or any implementation thereof.

[0078] In operation 502, the ions generated from the sample fraction eluted from column 110 accumulate over time t. accum Internal accumulation. Ions can accumulate in any device, such as ion accumulator 204 or mass analyzer 206 (e.g., a trapping device). Accumulation time t accum It can be initially set to the default or baseline cumulative time t base The default or baseline accumulation time is configured to generate a target accumulated ion population. The target accumulated ion population may be based on characteristics such as those of the ion accumulator 204, the mass analyzer 206, and / or the detector 208, such as storage space charge capacity and / or spectral space charge capacity.

[0079] In some instances, the cumulative time t accum The initial value can be manually set by the user (e.g., via controller 106 or controller 210). In another instance, the accumulated time t accum The initial value can be automatically set by the system 400, such as based on default values, method parameters used for a specific method or determination, characteristics of the ion accumulator 204, mass analyzer 206, or detector 208, and / or based on information provided by the user (e.g., based on the amount of the target analyte). The accumulation time t can be set. accum The initial value is set to prevent exceeding or falling below the target ion quantity.

[0080] In operation 504, a mass spectrum of the ions derived from the accumulated ions is acquired. Operation 504 can be acquired in any suitable manner. For example, if ions are accumulated in ion accumulator 204, the accumulated ions are transferred to mass analyzer 206, which performs mass analysis of the accumulated ions, such as a full-spectrum MS scan, or an MS / MS pre-scan or analytical scan, where the detected ions are product ions. If ions are accumulated in a trap-type mass analyzer, the mass analyzer performs mass analysis on the accumulated ions. The mass spectrum is then stored together with a series of mass spectra 506 previously acquired and stored during the experiment (e.g., in storage facility 402). If the mass spectrum is the first mass spectrum acquired during the experiment, the mass spectrum is stored, and subsequently acquired mass spectra can be combined with the mass spectrum to produce a series of mass spectra 506.

[0081] In operation 508, system 400 obtains an elution curve from mass spectrum 506, which includes multiple detection points, each from a different acquisition and representing the change in intensity of ions detected by mass spectrometer 104 over time. As mentioned, the intensity value at each detection point can represent the change in the corresponding total ion current contained in mass spectrum 506 over time.

[0082] In operation 510, system 400 determines whether the cumulative time setting conditions are met. The cumulative time setting conditions ensure that there is sufficient data for system 400 to set the cumulative time based on a subset of the detection points contained in the elution curve. In some instances, the cumulative time setting conditions include determining a minimum or predetermined number of detection points acquired (e.g., 6, 24, 48, 100, etc.).

[0083] In alternative examples, the cumulative time setting condition includes determining the minimum amount of time that has elapsed since a reference time. For example, the reference time could be, for instance, the time of the first acquisition, the time when sample 108 is injected into the liquid chromatograph 102, or the time when instrument conditions or method parameters change. In some instances, system 400 may determine whether the time elapsed since the reference time is greater than or equal to the size of a sliding time window. The sliding time window may cover periods such as 0.1 seconds, 0.5 seconds, 3 seconds, etc.

[0084] If the accumulated time setting condition is not met, the processing of method 500 returns to operation 502 for another fetch. However, if the accumulated time setting condition is met, the processing of method 500 proceeds to operation 512.

[0085] In operation 512, system 400 classifies the current signal state of the ion signal of the elution curve based on a subset of detection points included in a plurality of detection points of the elution curve and based on signal state model 514. The signal state indicates how the intensity signal currently changes over time at a detection point (e.g., the latest detection point) selected by system 400. For example, the signal state could be a baseline state indicating little change in detected intensity, an ascending state indicating an increase in detected intensity toward an estimated or predicted peak in the elution curve, a peak state indicating detected intensity near the peak peak, or a descending state indicating a decrease in detected intensity from the peak peak toward the baseline value. In some instances, the current signal state of the ion signal can be classified from a plurality of signal states (e.g., baseline state, ascending state, peak state, and descending state) based on the estimated position of the selected detection point (e.g., the latest detection point) relative to an expected reference point (e.g., the estimated or predicted peak peak). An illustrative embodiment of operation 512 will be described in more detail below.

[0086] In operation 516, system 400 sets the cumulative time t based on the current signal state classified in operation 512. accum For example, system 400 can determine the nominal accumulation time t based on the ion flux at a selected detection point (e.g., the most recent detection point). nom Then, the nominal cumulative time t is corrected using a correction factor associated with the current signal state. nom To obtain the cumulative time t accum An illustrative embodiment of operation 516 will now be described in more detail.

[0087] After operation 516, the processing of method 500 returns to operation 502 for another fetch, where the accumulated time t... accum This was previously set in operation 516. Method 500 can continue until it is terminated automatically or manually.

[0088] Now see Figure 6 The illustrative embodiment describing the execution of operation 512 is shown in the figure, illustrating an illustrative TIC 600 that can generate or represent a series of mass spectra 506, and see also [link to relevant documentation]. Figure 7 This illustrates TIC 600 after data from two additional acquisitions has been added to it. TIC 600 will be helpful in describing operation 512, but operation 512 can be performed with the original source data without generating TIC 600.

[0089] The TIC 600 plots multiple detection points 602, each representing a detected intensity value (in arbitrary units) varying with retention time (in minutes) detected by the mass spectrometer 104 during pre-scanning or analytical acquisition. Each consecutive mass spectrum acquired by the mass spectrometer 104 adds a new detection point 602 to the TIC 600. As shown on the TIC 600, the rightmost detection point 602-1 is at the current time t. c And it has a current intensity value I as indicated by dashed line 604. c The current (nearest) detection point is 602. (As in...) Figure 6 As shown, detection point 602 together forms the elution profile of the component eluted from column 110. The rising trend of the intensity value at detection point 602 indicates the start of peak 606 in the expected (e.g., estimated or predicted) elution profile. Figure 6 In the diagram, the expected elution curve is represented by a dashed curve.

[0090] Figure 8 It shows the Figure 6 or Figure 7 An illustrative method 800 (e.g., operation 512) for classifying the current signal state of the ion signals from the elution curve. Although Figure 8 The illustration shows a descriptive operation according to one embodiment, but other embodiments may omit, add, reorder, and / or modify it. Figure 8 Any of the operations shown. Figure 8 One or more of the operations shown can be performed by LC-MS system 100 and / or system 400, any components contained therein and / or any implementation thereof.

[0091] In operation 802, system 400 selects a specific detection point 602 in the elution curve. In the following example, the selected detection point 602 is the current detection point 602-1. In an alternative example, the selected detection point can be any historical detection point 602 acquired during the period (e.g., at the current time t). c Any previously acquired detection point 602, such as the second or third most recent detection point 602, or the selected detection point can be a combination of multiple historical detection points (e.g., the average value).

[0092] In operation 804, system 400 estimates the relative position of the selected detection point 602-1 relative to a reference point in the elution curve. The estimation of the relative position of the selected detection point 602-1 is based on the principle that the relative position of the selected detection point 602-1 varies with the detected intensity values ​​of multiple detection points 602 surrounding (e.g., before and / or after) the selected detection point 602-1. Therefore, the position of the selected detection point 602-1 relative to a reference point of peak 606 (e.g., the expected or predicted peak 608) can be estimated in real time and used to identify the current signal state of the ion signal in the elution curve.

[0093] In some instances, the relative position of the selected detection point 602-1 with respect to the reference point is the normalized intensity value of the selected detection point 602-1 relative to the reference intensity value. For example, in... Figure 6 and 7 The reference intensity value shown can be, for example, the expected maximum intensity value I at the apex 608 of peak 606. max As indicated by the dashed line 610. However, since it has not yet been reached at the current time t... c The maximum intensity value I of peak 606 was detected. max Therefore, the intensity values ​​of different sets 612 of multiple detection points 602 are used as input to the signal state model 514, which is configured to estimate the normalized intensity value of the selected detection point 602-1.

[0094] In some instances, set 612 includes a predetermined number (e.g., 6, 24, 48, 100, etc.) of detection points 602. In alternative instances, set 612 includes only the detection points 602 that appear within a sliding time window. For example, the sliding time window may cover periods of 0.1 seconds, 0.5 seconds, 3 seconds, etc. In either configuration, the selected detection points 602 are included in set 612. If the detection points 602 are not evenly spaced along the time axis, they can be corrected (e.g., by interpolation) to a fixed and uniform time interval (e.g., 1 second) to simplify processing.

[0095] Signal state model 514 is configured to use set 612 as input to perform any suitable heuristic steps, processes, and / or operations that can be performed or executed by system 400 to estimate the normalized intensity value of the selected detection point 602-1. In some instances, signal state model 514 may be implemented by hardware and / or software components (e.g., processor, memory, communication interface, instructions stored in memory for processor execution, etc.), such as storage facility 402 and / or processing facility 404 of system 400. Signal state model 514 may contain any suitable algorithm and / or machine learning model configured to estimate the normalized intensity value of the selected detection point based on the intensity values ​​of a set of historical detection points (e.g., set 612). Signal state model 514 may estimate the normalized intensity value in any suitable manner. In some instances, signal state model 514 includes a machine learning model. Illustrative machine learning models and methods for training machine learning models will be described in more detail below.

[0096] When the reference intensity value is the expected maximum intensity value I at the peak 608 of peak 606. max At that time, the normalized intensity value of the selected detection point 602-1 will typically be in the range of 0 to 1. Figure 6 In this example, system 400 estimates the current normalized intensity value of the selected detection point 602-1 to be approximately 0.15, and therefore indicates the current normalized intensity value I of the selected detection point 602-1. c The expected intensity value I at vertex 608 max Approximately 15%. Figure 7 In this example, system 400 estimates the current normalized intensity value I of the current detection point 602-1. c It will be approximately 0.65.

[0097] In the example described above, the reference intensity value is the expected maximum intensity value I at the apex 608 of peak 606. max However, any other normalization scheme can be used, and the reference intensity value can be any other suitable reference value, such as a known running average intensity value, the overall maximum intensity value of multiple different m / z values, the most recent maximum intensity value, etc.

[0098] In the example just described, the relative position of the selected detection point 602-1 is the normalized intensity value of the selected detection point 602-1 relative to the reference intensity value. In other examples, the relative position of the selected detection point 602-1 is the time distance of the selected detection point 602-1 relative to the time reference point. For example, as in Figure 6 and 7 As shown, the time reference point may appear at peak 606, where the intensity value is expected to reach its maximum value I at peak 608. max Expected time t maxSystem 400 can estimate the time distance (e.g., t) from the selected detection point 602-1 to the reference point in any suitable manner. max With t c (The difference between them).

[0099] In some instances, the time distance is estimated in a manner similar to estimating the normalized intensity value of the selected detection point 602-1. For example, the set 612 of detection points 602 and / or the estimated normalized intensity value are applied as input to a signal state model 514, which is additionally or alternatively configured to estimate the expected time t based on a set of historical detection points. max .like Figure 6 As shown, system 400 can estimate the time distance of selected detection points 602-1 to 0.05 minutes (e.g., 3 seconds) based on set 612. Figure 7 As shown, system 400 can estimate the time distance of selected detection point 602-1 to 0.02 minutes (e.g., 1.2 seconds) based on set 612.

[0100] Refer again Figure 8 In operation 806, system 400 identifies the signal state associated with the estimated relative position of the selected detection point 602-1. Operation 806 can be performed in any suitable manner. In some instances, system 400 can select a specific signal state from multiple signal states based on the relative position of the selected detection point 602-1 estimated in operation 804.

[0101] For example, the current signal state of the elution curve can have one of many different signal states. Figure 9 This illustrates the normalized intensity values ​​based on a reference intensity point relative to the elution curve (e.g., vertex 608). Figure 7 The elution curve of the TIC600 shown is divided into several distinct signal state regions 902 (e.g., signal state regions 902-1 to 902-4) as an example. For instance, before and after the detection of vertex 608, the baseline region 902-1 lies below a first threshold normalized intensity value 904-1 (e.g., 0.1). Before the detection of vertex 608, the rising region 902-2 lies between the first threshold normalized intensity value 904-1 and the second threshold normalized intensity value 904-2 (e.g., 0.8). Vertex region 902-3 lies above the second threshold normalized intensity value 904-2 and before and after the detection of vertex 608. The falling region 902-4 lies between the second threshold normalized intensity value 904-2 and the first threshold normalized intensity value 904-1 and after the detection of vertex 608. Although... Figure 9Four distinct signal state regions 902 are shown, but any other suitable configuration and number of signal state regions can be used. For example, rising region 902-2 and falling region 902-4 can be based on different threshold-normalized intensity values. Furthermore, more or fewer than four signal state regions can be used (e.g., baseline region 902-1, rising region 902-2, vertex region 902-3, and / or falling region 902-4 can each be divided into multiple distinct regions).

[0102] System 400 can identify the signal state associated with the estimated relative position of the selected detection point 602-1 by identifying the signal state region 902 where the selected detection point 602-1 is located. For example, system 400 can compare the normalized intensity value of the selected detection point 602-1 with a signal state boundary table representing the signal state region 902 to identify the current signal state associated with the selected detection point 602-1. Figure 9 In one instance, system 400 can determine that the selected detection point 602-1 is located within the rising region 902-2, and therefore classify the current signal state as a rising signal state. In some instances, system 400 can apply set 612 to signal state model 514, which classifies the selected detection point 602-1 according to the signal state region 902 in which the selected detection point 602-1 is located.

[0103] In other instances, the relative position of the selected detection point 602-1 is the time distance to a time reference point (e.g., the expected time t). max When the system 400 is in use, it can identify the current signal state of the ion signal based on the estimated relative position of the selected detection point 602-1. Figure 10 This shows a time reference point based on time and relative to the elution curve (e.g., expected time t). max )Will Figure 7 The elution curves of the TIC 600 shown are illustrated by dividing the data into different signal state regions 1002 (e.g., regions 1002-1 to 1002-4). Although Figure 10 Four different signal state regions 1002 are shown, but any other suitable number of signal state regions can be used.

[0104] As shown, at the expected time t max (For example, 0.6 minutes prior) the baseline region 1002-1 was located before the first threshold time distance 1004-1. At the expected time t... max (For example, 0.2 minutes) before, the rising region 1002-2 lies between the first threshold time distance 1004-1 and the second threshold time distance 1004-2. At the expected time t... max(For example, 0.2 minutes later), the vertex region 1002-3 lies between the second threshold time distance 1004-2 and the third threshold time distance 1004-3. At the expected time t... max (For example, 1.4 minutes later), the falling region 1002-4 lies between the third threshold time distance 1004-3 and the fourth threshold time distance 1004-4. The baseline region 1002-1 also lies after the fourth threshold time distance 1004-4. Figure 10 The configuration of signal state region 1002 shown is illustrative only, as any other configuration can be used. For example, any one or more signal state regions 1002 can be defined relative to another reference point, such as when the intensity value is expected to reach a minimum value I. min Expected time t min As indicated by dashed line 614.

[0105] System 400 can identify the signal state of the ion signal in the elution curve by recognizing the signal state region 1002 where the selected detection point 602-1 is located. For example, system 400 can compare the estimated time distance of the selected detection point 602-1 with a signal state boundary table representing the signal state region 1002 to identify the current signal state associated with the selected detection point 602-1. Figure 10 In one instance, system 400 can determine that the selected detection point 602-1 is located within the rising region 1002-2, and therefore classify the current signal state as a rising signal state. In some instances, system 400 can apply set 612 to signal state model 514, which classifies the selected detection point 602-1 according to the signal state region 1002 in which the selected detection point 602-1 is located.

[0106] As mentioned above, in some instances, the signal state model 514 includes a machine learning model configured to estimate the relative position of selected detection points with respect to a reference intensity and / or a time point, and to classify the current signal state of the ion signal of the elution curve based on the estimated relative position of the selected detection points. An illustrative method for training the machine learning model will now be described. Figure 11A method 1100 for training a machine learning model that can implement a signal state model 514 is illustrated. As shown, training data 1102 can be provided to a model training facility 1104, which can use the training data 1102 to train a signal state model 1106. The following example describes training the signal state model 1106 to estimate the normalized intensity value of a selected detection point and classify the current signal state of the ion signal of the elution curve based on the estimated normalized intensity value of the selected detection point. However, the same principle can be applied to training the signal state model to classify the current signal state based on the estimated time distance of the selected detection point from the time reference point.

[0107] Model training facility 1104 can perform any suitable heuristic steps, processes, and / or operations that can be configured to train a machine learning model. In some instances, model training facility 1104 may be implemented by hardware and / or software components such as storage facility 402 and / or processing facility 404 of system 400 (e.g., processor, memory, communication interface, instructions stored in memory for execution by the processor, etc.).

[0108] The signal state model 1106 can be any suitable type of machine learning model, such as a neural network model (e.g., a convolutional neural network (CNN)), a boosted decision tree regression model, a decision forest regression model, a fast forest quantile regression model, and an ordered regression model.

[0109] Training data can be obtained or extracted from data representing one or more elution curves (e.g., a set of LC-MS detection points) 1102. Figure 12 This shows a dataset that can be included in training data 1102. Figure 12 An illustrative TIC 1200 is shown, which generates or represents the LC-MS system from data acquired from the LC-MS system 100. It will be appreciated that a signal state model 1106 can be trained based on the TIC 1200 source data without generating the TIC 1200. As shown, the TIC 1200 plots multiple detection points 1202, each representing a detected intensity value, as detected by the mass spectrometer during full-spectrum acquisition and retention time (minutes). Each subsequent acquisition adds a new detection point 1202 to the TIC 1200. The detection points 1202 together form an elution curve 1204 of the component eluted from column 110. The elution curve 1204 contains a peak 1206 with a vertex 1208, at which the detected intensity is at its maximum intensity value I. max As indicated by the dashed line 1210.

[0110] The signal state model 1106 is trained based on the normalized intensity value or time distance of the selected detection point 1202 (e.g., the detected intensity value of the selected detection point 1202 is compared with that of I). max The principle is that the ratio of known reference intensity values ​​varies with the intensity levels detected from one or more historical detection points 1202 from the same experiment. Therefore, the training data 1102 applied to the model training facility 1104 includes a series of input vectors, each containing the detected intensity values ​​of a different set of detection points 1202. Each input vector can include any different set of detection points 1202 that can serve a particular implementation. For example, a first input vector can include the detected intensity values ​​of a first set 1212-1 of detection points 1202, a second input vector can include the intensity values ​​of a second set 1212-2 of detection points 1202, a third input vector can include the intensity values ​​of a third set 1212-3 of detection points 1202, and so on.

[0111] exist Figure 12 In this example, each input vector contains intensity values ​​for ten detection points 1202. However, each input vector can contain any other suitable number of detection points 1202 (e.g., 6, 12, 24, etc.). Alternatively, each input vector can contain intensity values ​​for all detection points 1202, which are contained in one or more previous input vectors.

[0112] In some instances, the detection points 1202 may not be evenly spaced along the time axis. To simplify the training of the signal state model 1106, the detection points 1202 can be corrected (e.g., by interpolation) to a fixed and uniform time interval (e.g., 1 second).

[0113] Any number of input vectors can be applied to the model training facility 1104. In some instances, the number of input vectors is chosen to at least cover the full half-width of peak 1206. Figure 12 In the example, each of the five input vectors with intensity values ​​at ten detection points will at least cover the full half-width of peak 1206. Alternatively, the number of input vectors can be chosen to cover the full width of peak 1206. Input vectors covering the full width of peak 1206 can be used to train signal state model 1106 to estimate the expected time t for the peak intensity values ​​to reach the baseline level. min .

[0114] For each input vector, a specific detection point 1202 is selected as the selected detection point. Any detection point 1202 contained within each corresponding input vector can be selected. In the following examples, the selected detection point 1202 for each input vector is the rightmost (nearest) detection point 1202 contained within the corresponding input vector. For example, the first input vector contains detection point 1202-1 as the selected detection point, the second input vector contains detection point 1202-2 as the selected detection point, and the third input vector contains detection point 1202-3 as the selected detection point. In alternative instances, the selected detection point 1202 is not the rightmost detection point, but can be any other suitable detection point (e.g., detection point 1202-1 is the selected detection point of the second input vector defined by set 1212-2, detection point 1202-2 is the selected detection point of the third input vector defined by set 1212-3, etc.).

[0115] The training data 1102 also contains known expected output values ​​from the signal state model 1106. The known expected output values ​​include normalized intensity values ​​(or temporal distance values ​​for each selected detection point 1202 per input vector) for each input vector. The normalized intensity values ​​are known because of the reference intensity value (e.g., I). max The output value is known. The known output value can be used for supervised training of the signal state model 1106.

[0116] To simplify the training of the signal state model 1106, detection points 1202 can be corrected based on known reference intensity values ​​in some instances. That is, the detected intensity values ​​of each detection point 1202 can be normalized based on known reference values.

[0117] In the example described above, the reference intensity value is the expected maximum intensity value I at the apex 1208 of peak 1206. max However, any other normalization scheme can be used, and the reference intensity value can be any other suitable reference value, such as a known running average intensity value, an overall maximum intensity value, a recent maximum intensity value, etc. In some instances, the normalization scheme may include baseline slope and / or baseline level corrections. For example, the normalized intensity value can be determined as the current intensity value I. c Compared with baseline intensity value I min The difference between them and the expected maximum intensity value I max Compared with baseline intensity value I min The ratio of the differences between ((I) c -I min ) / (I max -I min )).

[0118] In some instances, training data 1102 can be split into two sets of data, such that a first set of training data can be used to train signal state model 1106, and a second set of training data can be used to score signal state model 1106. For example, training data 1102 can be split such that a first percentage (e.g., 75%) of the input vectors can be used as the training set for training signal state model 1106, and a second percentage (e.g., 25%) of the input vectors can be used as the scoring set to generate an accuracy score for signal state model 1106.

[0119] During the training phase, model training facility 1104 can run one or more sessions based on training data 1102 to train signal state model 1106 to estimate the normalized intensity value of selected detection points 1202. Model training facility 1104 can also run one or more sessions based on training data 1102 to train signal state model 1106 to estimate the time distance from selected detection points to a reference time, such as the expected t when the intensity value of the expected elution curve reaches its maximum intensity value (e.g., the peak of the elution curve). max And / or the expected time t when the intensity value of the expected elution curve reaches the baseline intensity value. min The model training facility 1104 can also run one or more sessions based on the training data 1102 to train the signal state model 1106 to estimate the signal state region where the selected detection points are located. The model training facility 1104 can use any suitable machine learning technique or algorithm to perform operations to facilitate learning how to fit the machine learning model to the intensity values ​​detected within the first set of training data 1102.

[0120] Completing the training phase via model training facility 1104 results in a trained signal state model 1106 being configured to estimate the relative positions of selected detection points and classify the current signal state of the ion signal. The trained signal state model 1106 can be stored in a data store such as storage facility 402 and can be executed during runtime by any suitable computational component including processing facility 404.

[0121] If data from multiple different chromatographic conditions are obtained from a survey with a sufficiently short period to characterize the chromatographic peak (Nyquist limit), then the trained signal state model 1106 can be applied to experimental datasets that can use different chromatographic conditions and have different peak widths. The Nyquist limit of a Gaussian curve is six points. Therefore, if the training data 1102 contains six detection points spanning the chromatographic peak, interpolation over a fixed time interval between sampling points can be robust.

[0122] In an alternative instance, the signal state model 1106 can be trained based on training data 1102 configured for a specific application, such as a selected m / z range, specific chromatographic conditions, a specific sample type, etc. In such an instance, the signal state model 1106 can be trained after data from an initial start-up experiment is acquired, and the signal state model 1106 can be used thereafter only for subsequent iterations of the specific experiment.

[0123] In some instances, such as when the estimated normalized intensity value or the estimated time distance is found to deviate significantly from the actual value (e.g., by a predetermined amount) during the experiment, the signal state model 1106 can be improved or further trained in real time (e.g., during the experiment).

[0124] Now refer to Figure 13 Description of operation 516 of method 500 (see also) Figure 5 Illustrative examples of ). Figure 13 An illustrative method 1300 is shown for setting the cumulative time based on the current signal state of the ion signal according to the elution curve. Although Figure 13 The illustration shows a descriptive operation according to one embodiment, but other embodiments may omit, add, reorder, and / or modify it. Figure 13 Any of the operations shown. Figure 13 One or more of the operations shown can be performed by LC-MS system 100 and / or system 400, any components contained therein and / or any implementation thereof.

[0125] In operation 1302, system 400 determines the nominal cumulative time t based on the ion flux at the selected detection point. nom Then, the nominal cumulative time t is corrected using a correction factor associated with the current signal state of the ion signal from the elution curve. nom To obtain the cumulative time t accum Nominal cumulative time t nom It can be determined by the following equation (1):

[0126]

[0127] Where t nom It is the nominal accumulation time for the next acquisition, the target is the number of target ions to be accumulated in the next acquisition, and the ion flux is the ion flux (in ions per unit time) measured at the selected detection point. As explained above, the target is previously determined based on device characteristics (e.g., space charge capacity), experimental conditions, and / or method parameters.

[0128] In operation 1304, system 400 adjusts the nominal cumulative time t according to the following equation (2) by using a correction factor C(ss) associated with the current signal state (ss) classified in operation 512. nom Determine the cumulative time t accum :

[0129] t accum =t nom *C(ss) (2)

[0130] Where C(ss) is a correction factor associated with the current signal state, such as the baseline state correction factor C. base Rising state correction factor C rising Vertex state correction factor or descent state correction factor C falling Each correction factor C(ss) is set to prevent the number of ions to be accumulated from exceeding or falling below the target number. For example, the baseline state correction factor C... base It can be approximately 1.0, the rising state correction factor can be less than 1.0, and the vertex state correction factor C apex It can be approximately 1.0, and the falling state correction factor C falling It can be greater than 1.0.

[0131] In operation 1306, system 400 sets the accumulation time of the device (e.g., ion accumulator 204 and / or mass analyzer 206) to the accumulation time t determined in operation 1304. accum .

[0132] like Figure 14 Another illustrative method 1400 is shown for setting the accumulation time based on the current signal state of the ion signal from the elution curve. Although Figure 14 The illustration shows a descriptive operation according to one embodiment, but other embodiments may omit, add, reorder, and / or modify it. Figure 14 Any of the operations shown. Figure 14 One or more of the operations shown can be performed by LC-MS system 100 and / or system 400, any components contained therein, and / or any implementation thereof. Method 1400 is similar to method 1300, except that in method 1400, the cumulative time t during the baseline state is... accum No adjustment is made using the correction factor C(ss).

[0133] In operation 1402, system 400 determines the nominal cumulative time t. nom Operation 1402 can be performed in any suitable manner, such as as described for operation 1302.

[0134] In operation 1404, system 400 determines in operation 512 (see...) Figure 5 Check whether the current signal state, as classified in the previous step, is the baseline signal state. If the current signal state is not the baseline signal state, then proceed to operations 1406 and 1408.

[0135] In operations 1406 and 1408, system 400 determines the cumulative time t. accum Set the device's cumulative time to the cumulative time t. accum Operations 1406 and 1408 can be performed in any suitable manner, such as as described for operations 1304 and 1306 respectively.

[0136] If the current signal state is the baseline signal state, the processing proceeds to operation 1410. In operation 1410, system 400 determines the constant baseline accumulation time t. base (e.g., 30 milliseconds) and nominal cumulative time t nom The minimum value of is given by the following equation (3):

[0137] t accum =Time(t) base ,t nom (3)

[0138] As explained above, the baseline cumulative time t base It can be determined in advance and configured to generate a targeted cumulative ion population. Baseline accumulation time t base It can be determined based on experience, manually, or through optimization, as will be described in more detail below.

[0139] In operation 1412, the cumulative time t accum Set as baseline cumulative time t base and nominal cumulative time t nom The minimum value. Accumulated time t during the baseline state. accum Set as baseline cumulative time t base and nominal cumulative time t nom The minimum value provides a balance between sensitivity (e.g., ensuring sufficient ion accumulation) and not exceeding the target ion count. This is due to the baseline accumulation time t. base The initial ion flux is adjusted, thus allowing the baseline accumulation time t to be set. base To manage or prevent overshoot, for example, if the current signal state is incorrectly classified as the baseline state when the elution curve is actually rising. In this case, the baseline state is set to be less than the nominal accumulation time t. nom This can prevent (or at least reduce) exceeding the target.

[0140] The correction factor C(ss) associated with each signal state and used in methods 1300 and 1400 can be determined in any suitable manner. In some instances, each correction factor C(ss) is determined empirically, iteratively, or through a suitable algorithm (e.g., optimization algorithm, machine learning model, neural network, etc.). In some instances, as will now be referred to... Figure 15 and 16 The correction factor C(ss) described is determined based on the principle of reinforcement learning, which borrows from the fact that the action to be taken by the agent is determined by the reward policy given a specific state. Here, the agent is system 400, and the action to be taken is set to a cumulative time t. accum And the state is the current signal state of the ion signal in the elution curve.

[0141] Figure 15 Table 1500 illustrates an illustrative reward strategy 1502 that can be used to determine the correction factor C(ss) for each signal state. Table 1500 plots the reward R as a function of the ion / target ratio (the ratio of the actual number of accumulated ions to the target number of accumulated ions). In this example, the reward R is assigned a value ranging from -1.5 to 1.0 based on the value of the ion / target ratio, with the maximum reward (value 1.0) occurring when the ion / target ratio is 1.0 (e.g., when the actual number of accumulated ions is the same as the target number of accumulated ions). However, the reward R can be assigned any other value that may be suitable for a particular implementation.

[0142] As in Figure 15 As observed, when the ion / target ratio is greater than 1.0, the slope of the reward strategy 1502 is steeper than when the ion / target ratio is less than 1.0. This is because accumulating too many ions leads to space charge effects, such as mass transfer, peak broadening, and coalescence, thus penalizing exceeding the target more than penalizing falling below the target. For example, for the reward strategy 1502, falling below the target by 25% results in a reward of approximately 0.75, while exceeding the target by 25% results in a reward of approximately 0.4. It will be appreciated that the reward strategy 1502 (e.g., the shape, slope, value, etc. of the reward R) can be modified as desired and can be tailored to specific instrument properties (e.g., the ion capacity of ion accumulator 204 or mass analyzer 206), specific measurements (e.g., the amount of target analyte, the shape of the peaks in the elution curve (e.g., Gaussian, non-Gaussian, etc.)), and / or experimental conditions. Typically, the reward strategy 1502 is based on the characteristics of the measurement system, such as the cost of overfilling the analyzer.

[0143] In some embodiments, a separate validation metric can be used to optimize the reward strategy 1502. For example, a validation metric could be determining the dynamic quantitative range of a range of analytes using experimental or simulated data. This validation metric can be used to inform decisions regarding the shape and configuration of the reward strategy 1502.

[0144] The reward strategy 1502 can be used to determine the value of the correction factor C(ss). As explained above, the actual number of accumulated ions is based on the ion flux and the accumulation time t. accum This depends on the correction factor C(ss), as shown by equation (2) above. The correction factor C(ss) can be determined by optimizing the return (as determined by return strategy 1502) over the entire elution curve and therefore over all states. A set of correction factors C(ss) can be obtained (e.g., C0). base C rising C apex and C falling The set of factors is used to optimize returns, and the correction factor results in the best returns (e.g., best total return, best average return, etc.) across the entire elution curve or the portion of the elution curve of interest.

[0145] As used herein, “optimization” and its variations mean searching for an improved or optimal solution within a set of possible solutions, but may not necessarily yield the optimal solution, such as when the optimization process terminates before finding the optimal solution, when multiple solutions exist that satisfy predefined criteria, when a solution satisfies a minimum criterion, or when the selected optimization technique fails to converge to the optimal solution. Similarly, as used herein, an “optimal” parameter (e.g., the “maximum” or “minimum” value of a parameter) means a solution obtained by performing an optimization process, and therefore is not necessarily an absolute extremum of the parameter (e.g., an absolute maximum or minimum value), but still adjusts the parameter and leads to improvement.

[0146] In some instances, the reward strategy 1502 can be represented by a mathematical formula that can be used to optimize the reward using standard optimization techniques. In other instances, machine learning models (e.g., neural networks) or optimization algorithms (e.g., genetic algorithms) can be used to optimize the reward, such as by performing multidimensional optimization to simultaneously optimize multiple correction factors C(ss).

[0147] In another approach, the reward is optimized via one-dimensional optimization (e.g., 1D linear search), where the optimal value of each correction factor C(ss) is determined independently of other correction factors C(ss). The optimal value of a specific correction factor C(ss) (e.g., the optimal value of correction factor C) risingThe value of the correction factor C(ss) that produces the best return across the entire elution curve is given. To determine the optimal value of a specific correction factor C(ss) using this method, multiple fragments are run. Each fragment is run by acquiring a series of mass spectra using the method 500 described above and a specific combination of correction factors C(ss), as performed by executing an analytical scan. The specific correction factor C(ss) to be optimized varies across the different fragments, while other correction factors C(ss) remain constant across all fragments. The constant values ​​of other correction factors C(ss) can be previously estimated or determined using this 1D linear search optimization method. For each fragment, all acquired returns R are determined using return strategy 1502 and summed to produce the total return. The optimal value of the specific correction factor C(ss) is the value that produces the best total return for the elution curve. This method for optimizing the return can be expressed by the following equation (4):

[0148]

[0149] Where R(opt) is the reward for optimization, and R(ss) is the reward for optimization. t ) is the reward obtained each time in the segment determined by the reward strategy 1502, which varies with a specific signal state ss at each time t (and therefore a specific correction factor C(ss)).

[0150] refer to Figure 16 This explains C rising Determining the optimal value. Figure 16 An illustrative figure 1600 is shown, which plots the total return of multiple segments as a function of the correction factor C. rising Variation. Figure 1600 contains multiple points 1602, each point representing a specific segment of a specific set with a correction factor. In this example, by applying the correction factor C to all segments... base C apex and C falling Each segment remains at a constant value, while the correction factor C for each segment is changed. rising To generate point 1602, the correction factor C can be manually or iteratively adjusted according to the algorithm that converges to a solution. rising .exist Figure 16 The example uses a minimization algorithm, so the total reward for each segment is multiplied by -1 (or the reward strategy 1502 could be reversed to make the optimal reward -1.0). However, a maximization algorithm can be used without multiplying the total reward by -1. For example, in... Figure 16 As shown, in this example, when the correction factor C rising The minimum return occurs when the value is 0.354. Then, when running other segments, the correction factor C can be used. rising This value is used to independently optimize each of the other correction factors C(ss).

[0151] As mentioned above, in the example of method 1400, when the current signal state is the baseline state, system 400 will accumulate time t. accum Set as constant baseline cumulative time t base and nominal cumulative time t nom The minimum value, as given by equation (3) above, is not obtained by using the baseline correction factor C. base To correct the nominal cumulative time t nom As given in equation (2) above. In some instances, the baseline cumulative time t base The returns are determined by optimizing the correction factor C(ss) in a manner similar to that described above. For example, the baseline cumulative time t in equation (3) can be used. base Different values ​​of the correction factor C in equation (2) are used to run multiple segments, but the correction factor C in equation (2) is different. rising C apex and C falling The constant value can be estimated or determined prior. In some instances, due to the baseline cumulative time t base Adjusting ion flux and thus affecting the correction factor C rising C apex and C falling The value of , therefore, in optimizing the correction factor C rising C apex and C falling Previously optimized baseline cumulative time t base .

[0152] In another instance, the baseline cumulative time t base The number of target analytes in the experiment is used to determine the baseline cumulative time t. For example, system 400 can obtain a list or number of target analytes from a method file (which may be provided by the user) and set the baseline cumulative time t based on the number of target analytes. base To ensure cumulative time t accum Short enough to analyze all target analytes. The system 400 can also have the baseline accumulation time t set separately. base To ensure cumulative time t accum Long enough to achieve sufficient sensitivity.

[0153] In the example above, the return is optimized by finding the best total return across the entire segment, according to equation (4). However, the return can be optimized in any other suitable way, such as by finding the best average return across the segment.

[0154] The systems and methods for ion population modulation described above utilize information about peak shape and its intensity distribution previously acquired during one or more training experiments. For example, training experiments can be performed to train a signal state model 514, optimize the shape of the reward policy 1502, and optimize the reward given by the reward policy 1502 to determine the optimal correction factor C(ss) and / or the optimal baseline accumulation time t. base .

[0155] In some instances, the signal state model 514, the reward strategy 1502, and / or the correction factor C(ss) can be trained and / or optimized for various conditions. For example, by understanding information about the expected or predicted elution curve, the sampling frequency can also be optimized, such as by selecting a sampling frequency just above the Nyquist limit for the currently detected peak. The signal state model 514, the reward strategy 1502, and the correction factor C(ss) are trained on experimental data to handle various peak shapes. In another instance, the signal state model 514 can be trained to classify the current peak shape (e.g., Gaussian, non-Gaussian, etc.) and use peak shape information to select an appropriate set of correction factors associated with said specific peak shape. In yet another instance, the signal state model 514, the reward strategy 1502, and / or the correction factor C(ss) can be trained for specific experimental conditions, such as specific flow rates (e.g., nanoflow, microflow, flow range, etc.), specific measurements, target quantities of analytes, etc. In another instance, the signal state model 514 can be applied and trained on actual intensity values ​​rather than normalized intensity values.

[0156] In some instances, training experiments can be eliminated or reduced based on certain assumptions. For example, the signal state model 514 can be trained and optimized using simulated data on the average peaks in the elution curve, such as Gaussian peaks.

[0157] Signal state model 514, reward strategy 1502, correction factor C(ss) and / or baseline cumulative time t base The system can also score and / or update in real time during the analysis experiment based on data acquired during the analysis experiment (e.g., based on analysis acquisition or scanning). At different times throughout the analysis experiment, the system 400 can use the analysis data acquired up to that point to perform evaluations to assess the performance of the ion population regulation process (e.g., signal state model 514, reward strategy 1502, correction factor C(ss), and / or baseline cumulative time t). baseEvaluations can be performed at any suitable time, such as periodically (e.g., every nth scan), randomly, or in response to a triggering event (e.g., an event indicating overfilling, such as detecting coalescence or peak broadening exceeding a threshold amount). Each evaluation can assess the signal state model 514, the reward strategy 1502, the correction factor C(ss), and / or the baseline cumulative time t. base The quality.

[0158] For example, the evaluation can be conducted by comparing the actual peak of the elution curve to assess whether any detection points have been misclassified (based on the estimated relative position of the detection points). If the number and / or degree of misclassification exceeds a threshold, the system 400 can retrain the signal state model 514 using the acquired experimental data.

[0159] As another example, system 400 can determine how close the current correction factor C(ss) is to the return that would be achieved through a “perfect” correction factor C(ss) and / or a perfect return strategy. To do this, system 400 can run a segment with the currently acquired data to optimize the return of the current return strategy and use the obtained information to determine whether the return can be improved by changing the correction factor C(ss), the baseline cumulative time t, etc. base and / or return strategies to improve returns. If the currently applied correction factor C(ss), baseline cumulative time t base If the return deviates from the threshold by a certain amount from the "perfect" value, then System 400 can re-optimize the correction factor C(ss) and / or the baseline cumulative time t. base And / or select a better reward strategy for use during analytical experiments.

[0160] If the signal state model 514, the reward strategy 1502, the correction factor C(ss), and / or the baseline cumulative time t base If the parameters are updated during the analysis experiment, the updated parameters can be used as the default parameters for the next experiment. In this way, the ion population control system and method can be retrained and optimized in real time without spending additional time on retraining and optimization. The real-time evaluation and update process does not require user input, thereby improving user convenience.

[0161] In some instances, analytical experiments can begin without cumulative time correction or adjustment, using conventional methods of ion population regulation, or analytical experiments can use signal state models and parameters (e.g., reward strategy, correction factor C(ss), and / or baseline cumulative time t) determined previously from other previously acquired data (e.g., training data, analytical experiments, simulation data, etc.). baseDuring the analytical experiment, the signal state model and parameters for a specific analytical experiment can be trained and determined in real time using data acquired in the first phase of the analytical experiment. In the second phase of the analytical experiment, system 400 can implement method 500, which includes the trained signal state model and determined parameters, to regulate the ion population. The second phase can begin after thresholds such as the minimum number of peak elutions, the shortest elapsed time, the detection of a specific analyte, the minimum total ion current, etc., are reached. The initial data used to train the signal state model and determine the parameters to be applied, including data acquired in the first phase of the analytical experiment, may be small but available and can be refined in subsequent re-optimization, as described above. In some instances, the signal state model and parameters can be refined and re-optimized based on data acquired during the second phase of the analytical experiment. If the corrections or parameters applied during the second phase fall outside some threshold boundaries (e.g., acceptable tolerances), system 400 can revert to conventional methods using ion population regulation or use the signal state model trained on other data not previously acquired during the analytical experiment.

[0162] Various modifications can be made to the systems and methods described herein without departing from the scope and principles of the concepts described herein. For example, in the example described above, the training of the signal state model and the estimation of the relative positions of selected detection points are based on multiple detection points (e.g., TIC). In some modifications, the training of the signal state model, the estimation of the relative positions of selected detection points, and / or the optimization of the reward can be based on multiple m / z values ​​or a selected m / z (e.g., XIC) and / or a signal smaller than the entire elution curve (e.g., a small time window of interest).

[0163] In some instances, the system 400 can be configured to accumulate time t. accum Instead of determining the nominal cumulative time t nom The nominal cumulative time t is corrected using a correction factor C(ss). nom For example, system 400 can accumulate time t. accum Set to a cumulative time t similar to a constant baseline base The constant cumulative time value associated with each signal state, such as the rise cumulative time t. rising Vertex accumulation time and descent accumulation time t falling Each of these constant cumulative times can be determined in any suitable manner, such as empirically, manually, and / or by optimizing the reward strategy, as described above regarding the baseline cumulative time t. base The explanation given.

[0164] In some instances, the elution curve may exhibit different characteristics at different times, and therefore can be correlated with different payoff strategies and / or correction factors C(ss) or baseline cumulative time t at different times. base Related factors include, for example, different return strategies, correction factors C(ss), and baseline cumulative time t. base And / or the sampling frequency can be arranged to be used at different times during the elution curve, or it can be triggered when certain signals are detected (e.g., certain peak intensities, peak shapes, etc.).

[0165] In other instances, the principles described herein can be applied to the conditioning of ion populations in transport-type instruments rather than in trapping-type instruments where ions are accumulated. In these instances, the term "implantation time" refers to the time period during which ions are implanted into a device, such as a ToF mass filter or a quadrupole mass filter, and can be used interchangeably with accumulation time. Some transport-type instruments can additionally or alternatively use ion attenuation to modulate the intensity of the beam impacting the detector, and thereby modulate the ion population. In these instances, the degree of ion attenuation can be optimized in a manner similar to optimizing the implantation time, and can be optimized independently or in conjunction with the optimization of the implantation time.

[0166] In another modification, the separation apparatus (e.g., liquid chromatograph, gas chromatograph, capillary electrophoresis apparatus, etc.) and / or mass spectrometer (e.g., mass spectrometer 104) may include or be coupled to an ion mobility analyzer, and the data acquired by the ion mobility analyzer can be used to train a signal state model and estimate the relative positions of selected detection points in a manner similar to the methods described above for acquiring data via a mass spectrometer. For example, a first set of data acquired using the ion mobility analyzer and the mass analyzer may include a series of mass spectra containing ion intensity values ​​generated by the sample assembly that vary with the m / z and / or ion mobility (e.g., ion collision cross section (CCS)). A second set of data can be extracted from the first set of data. The extracted second set of data may include multiple detection points representing the intensity detected by the mass analyzer that varies with time at a selected m / z and / or with a selected CCS or range of CCS. The second set of data can be used in any of the ways described herein, such as training a signal state model and / or estimating the relative positions of selected detection points for a selected CCS or range of CCS.

[0167] In some instances, system 400 can be configured to request user input to manage or adjust settings for the ion population control process. For example, system 400 may obtain from the user method settings, a list of target analytes, the selected m / z, and / or any other initial or default values ​​for parameters associated with signal state model 514 and reward strategy 1502. System 400 can also be configured to notify the user of certain changes, such as when the cumulative time t... accum When the threshold value changes, or when changes are required, such as when the evaluation indicator parameter should be adjusted.

[0168] In some embodiments, one or more of the systems, components, and / or processes described herein may be implemented and / or performed by one or more suitably configured computing devices. For this purpose, one or more of the aforementioned systems and / or components may include, or be implemented by, any computer hardware and / or computer implementation instructions (e.g., software) embodied on at least one non-transitory computer-readable medium configured to perform one or more of the processes described herein. Specifically, system components may be implemented on one physical computing device or on more than one physical computing device. Therefore, system components may comprise any number of computing devices and may employ any number of computer operating systems.

[0169] In some embodiments, one or more processes described herein may be implemented at least in part as instructions embodied in a non-transitory computer-readable medium and executable by one or more computing devices. Typically, a processor (e.g., a microprocessor) receives instructions from a non-transitory computer-readable medium (e.g., memory, etc.) and executes those instructions, thereby performing one or more processes that include one or more processes described herein. Such instructions can be stored and / or transmitted using any of a variety of known computer-readable media.

[0170] Computer-readable media (also known as processor-readable media) include any non-transitory medium that participates in providing data (e.g., instructions) that can be read by a computer (e.g., by the computer's processor). Such media can take many forms, including, but not limited to, non-volatile and / or volatile media. Non-volatile media can include, for example, optical discs or magnetic disks and other permanent storage. Volatile media can include, for example, dynamic random access memory (“DRAM”), which typically constitutes main memory. Common forms of computer-readable media include, for example, magnetic disks, hard disks, magnetic tapes, any other magnetic media, optical disc read-only memory (“CD-ROM”), digital video discs (“DVD”), any other optical media, random access memory (“RAM”), programmable read-only memory (“PROM”), electrically erasable programmable read-only memory (“EPROM”), FLASH-EEPROM, any other memory chip or cassette disk, or any other tangible medium that a computer can read.

[0171] Figure 17 An illustrative computing device 1700 is shown, which can be specifically configured to perform one or more processes described herein. (As in...) Figure 17 As shown, the computing device 1700 may include a communication interface 1702, a processor 1704, a storage device 1706, and an input / output (“I / O”) module 1708 that are communicatively connected to each other via a communication infrastructure 1710. Although Figure 17 The illustration shows a computing device 1700, but... Figure 17 The components shown are not intended to be limiting. Additional or alternative components may be used in other embodiments. A more detailed description will now follow. Figure 17 Components of the computing device 1700 shown.

[0172] Communication interface 1702 can be configured to communicate with one or more computing devices. Examples of communication interface 1702 include, but are not limited to, wired network interfaces (such as network interface cards), wireless network interfaces (such as wireless network cards), modems, audio / video connections, and any other suitable interfaces.

[0173] Processor 1704 generally represents any type or form of processing unit capable of processing data and / or interpreting, executing, and / or directing the execution of one or more of the instructions, procedures, and / or operations described herein. Processor 1704 may perform operations by executing computer-executable instructions 1712 (e.g., applications, software, code, and / or other executable data instances) stored in storage device 1706.

[0174] Storage device 1706 may include one or more data storage media, devices, or configurations, and may employ data storage media and / or devices of any type, form, and combination. For example, storage device 1706 may include, but is not limited to, any combination of non-volatile media and / or volatile media described herein. Electronic data containing the data described herein may be stored temporarily and / or permanently in storage device 1706. For example, data representing computer-executable instructions 1712 configured to direct processor 1704 to perform any of the operations described herein may be stored within storage device 1706. In some instances, data may be arranged in one or more databases residing within storage device 1706.

[0175] I / O module 1708 may include one or more I / O modules configured to receive user input and provide user output. One or more I / O modules can be used to receive input for a single virtual experience. I / O module 1708 may include any hardware, firmware, software, or a combination thereof that supports input and output capabilities. For example, I / O module 1708 may include hardware and / or software for capturing user input, including, but not limited to, a keyboard or keypad, a touchscreen component (e.g., a touchscreen display), a receiver (e.g., an RF or infrared receiver), a motion sensor, and / or one or more input buttons.

[0176] I / O module 1708 may include one or more means for presenting output to a user, including, but not limited to, a graphics engine, a display (e.g., a screen), one or more output drivers (e.g., display drivers), one or more audio speakers, and one or more audio drivers. In some embodiments, I / O module 1708 is configured to provide graphical data to the display for presentation to the user. The graphical data may represent one or more graphical user interfaces and / or any other graphical content that may serve a particular implementation.

[0177] In some instances, any of the systems, computing devices, and / or other components described herein may be implemented by computing device 1700. For example, storage facility 402 may be implemented by storage device 1706, and processing facility 404 may be implemented by processor 1704.

[0178] Those skilled in the art will recognize that, although various illustrative embodiments have been described with reference to the accompanying drawings in the foregoing description, it will be apparent that various modifications and changes can be made thereto, and other embodiments can be implemented, without departing from the scope of the invention as set forth in the appended claims. For example, certain features of one embodiment described herein may be combined with or substituted for features of another embodiment described herein. Therefore, the description and drawings should be considered illustrative rather than restrictive.

Claims

1. A method for performing mass spectrometry, the method comprising: Ions generated by the components eluted from the chromatographic column accumulate over the accumulation period; The accumulated ions are transferred to a mass analyzer; During the acquisition process, mass spectra of the detected ions originating from the transferred ions are acquired; Elution curves are obtained from a series of acquired mass spectra, including the acquired mass spectra and multiple previously acquired mass spectra, the elution curves including multiple detection points, the multiple detection points representing the change in the intensity of the detected ions over time; Based on a subset of the detection points included in the plurality of detection points, the current signal state of the elution curve is classified into one of the groups including baseline state, rising state, peak state, and falling state. as well as The current signal state, classified based on the elution curve, is used to set the accumulation time for the next acquisition of the mass spectrum.

2. The method of claim 1, wherein obtaining the mass spectrum includes performing a pre-scan.

3. The method according to claim 1, wherein: The acquisition of the mass spectrum includes performing tandem mass spectrometry; and The detected ions include product ions.

4. The method according to any one of claims 1 to 3, wherein the elution curve is based on the total ion current of each mass spectrum contained in the series of acquired mass spectra.

5. The method of claim 1, wherein each signal state indicates how the intensity of the detected ion currently changes over time.

6. The method of claim 1, wherein setting the cumulative time based on the classified current signal state of the elution curve comprises: Determine the nominal cumulative time; Identify correction factors associated with the classified current signal state of the elution curve; as well as The nominal cumulative time is corrected based on the correction factor associated with the classified current signal state of the elution curve.

7. The method of claim 6, wherein the nominal cumulative time is determined based on the target ion population divided by the current ion flux.

8. The method of claim 6, wherein the correction factor is configured to optimize the returns given by the reward strategy.

9. The method of claim 6, wherein the correction factor is determined at least in part based on data obtained from the series of acquired mass spectra.

10. The method according to any one of claims 1 to 3, wherein the current signal state of the elution curve is classified based on the relative position of selected detection points included in the plurality of detection points, the relative position of the selected detection points representing the position of the selected detection points relative to a desired reference point in the elution curve.

11. The method of claim 10, wherein the selected detection point is the latest detection point.

12. The method of claim 10, wherein the relative position of the selected detection point includes a normalized intensity value of the selected detection point, the normalized intensity value representing the ratio of the intensity value detected at the selected detection point to the expected maximum intensity value of the elution curve.

13. The method of claim 10, wherein the relative position of the selected detection point includes the time distance of the selected detection point from the expected time point in the elution curve.

14. The method of claim 10, wherein classifying the current signal state of the elution curve comprises applying the subset of detection points to a trained machine learning model, the trained machine learning model being configured to estimate the relative position of the selected detection points based on the subset of detection points.

15. The method of claim 14, wherein the machine learning model is trained at least in part based on the plurality of detection points of the elution curve.

16. An apparatus for performing mass spectrometry, the apparatus comprising: An ion accumulator, configured to accumulate ions generated by components eluted from a chromatographic column during an accumulation time; A mass analyzer configured to acquire a series of mass spectra of detected ions originating from the accumulated ions during acquisition, after the accumulated ions have been transferred from the ion accumulator. as well as Computing device, the computing device being configured to: Elution curves are obtained from the series of mass spectra including the acquired mass spectra and multiple previously acquired mass spectra, the elution curves including multiple detection points representing the change in the intensity of the detected ions over time; Based on a subset of the detection points included in the plurality of detection points, the current signal state of the elution curve is classified into one of the following groups: baseline state, rising state, peak state, and falling state; and The current signal state, classified based on the elution curve, is used to set the accumulation time for the next acquisition of the mass spectrum.

17. The apparatus of claim 16, wherein the mass analyzer is configured to acquire the mass spectrum by performing a pre-scan.

18. The apparatus according to claim 16, wherein: The mass analyzer includes a tandem mass spectrometer; and The detected ions include product ions.

19. The apparatus of claim 16, wherein the elution curve is based on the total ion current of each mass spectrum contained in the series of mass spectra.

20. The device of claim 16, wherein each signal state indicates how the intensity of the detected ion currently changes over time.

21. The device according to any one of claims 16 to 19, wherein the computing device is configured to set the cumulative time based on the classified current signal state of the elution curve by: Determine the nominal cumulative time; Identify correction factors associated with the classified current signal state of the elution curve; and The nominal cumulative time is corrected based on the correction factor associated with the classified current signal state of the elution curve.

22. The device of claim 21, wherein the nominal cumulative time is determined based on the target ion population divided by the current ion flux.

23. The device of claim 21, wherein the correction factor is configured to optimize the reward given by the reward strategy.

24. The device of claim 21, wherein the correction factor is determined at least in part based on data obtained from the series of mass spectra.

25. The apparatus according to any one of claims 16 to 19, wherein the current signal state of the elution curve is classified based on the relative position of selected detection points included in the plurality of detection points, the relative position of the selected detection points representing the position of the selected detection points relative to a desired reference point.

26. The device of claim 25, wherein the selected detection point is the latest detection point.

27. The apparatus of claim 25, wherein the relative position of the selected detection point includes a normalized intensity value of the selected detection point, the normalized intensity value representing the ratio of the intensity value detected at the selected detection point to the expected maximum intensity value of the elution curve.

28. The apparatus of claim 25, wherein the relative position of the selected detection point includes the time distance of the selected detection point from the expected time point in the elution curve.

29. The apparatus of claim 25, wherein the computing device is configured to classify the current signal state of the elution curve by applying the subset of detection points to a trained machine learning model, the trained machine learning model being configured to estimate the relative position of the selected detection points based on the subset of detection points.

30. The apparatus of claim 29, wherein the machine learning model is trained at least in part based on the plurality of detection points of the elution curve.

31. A method comprising: Elution curves are obtained from a series of mass spectra including the latest mass spectrum and multiple previously acquired mass spectra, the elution curves including multiple detection points, each detection point representing the change in the intensity of ions accumulated over a cumulative time and detected by the detector over time. Based on a subset of the detection points included in the plurality of detection points, the current signal state of the elution curve is classified into one of the groups including baseline state, rising state, peak state, and falling state. as well as The current signal state, classified based on the elution curve, is used to set the accumulation time for the next acquisition of the mass spectrum.

Images