Qualitative analysis of mass spectrometry data

Abstract

A method of determining a confidence measure of precursor inference in mass spectrum data. The method includes evaluating the response profile using response profile metrics and reporting the confidence measure of the precursor inference using the response profile and the evaluating of the response profile. Also a method of detecting conflicting relationships between precursors including filtering an intensity at at least one position across the scanning ion transmission window dimension. Another method a method of detecting conflicting relationships between precursors includes generating, based on an encoding, a trace of the scanning ion transmission window dimension wherein each sum is substituted with a binary metric for presence of the at least one unique product ion in each window. Further a method of identifying a precursor m / z including determining an intercept between the rising linear function and the falling linear function, and identifying the precursor m / z using the intercept.

Description

Attorney Docket No. 18768.0109WOU1 / ABS-0840QUALITATIVE ANALYSIS OF MASS SPECTROMETRY DATACROSS-REFERENCE TO RELATED APPLICATION

[0001] This application is being filed as a PCT International Patent Application and claims priority to US Provisional Patent Application No. 63 / 692,907, filed September 10, 2024, entitled “QUALITATIVE ANALYSIS OF COMPRESSED MASS SPECTROMETRY DATA,” which is incorporated herein by reference in its entirety.BACKGROUND

[0002] Mass spectrometry (MS) is an analytical technique used to measure the mass- to-charge ratio of ions. It provides information about the molecular weight and structure of compounds by ionizing a sample, separating the ions based on their m / z ratios, and detecting them to generate a mass spectrum. In some variants of data-independent acquisition (DIA) method, the mass spectrometer collects data in segmented “windows” of m / z values.

[0003] In a DIA method, a precursor ion mass range is selected. A precursor ion transmission window is then stepped across the precursor ion mass range. All precursor ions in the precursor ion transmission window are fragmented and all of the product ions of all of the precursor ions in the precursor ion transmission window are mass analyzed. Each window represents a specific range of m / z values, and the instrument sequentially moves through these windows to cover a broader range of m / z values.

[0004] Some acquisitions use a relatively wide precursor ion transmission window, which is stepped across the entire precursor mass range. The precursor ion transmission window stepped across the precursor mass range in each cycle may have a width of 5-25 amu, or even larger. Use of the relatively wide precursor ion transmission window allows cycle time to be significantly reduced in comparison to the cycle time of methods using a narrower precursor ion transmission window.

[0005] Other DIA methods use a precursor ion transmission window scanned across a mass range so that successive windows have areas of overlap and non-overlap. This scanning makes the resulting product ions a function of the scanned precursor ion transmission windows. This additional information, in turn, can be used to identify the one or more precursor ions responsible for each product ion.

[0006] The correlation is done by first plotting the mass-to-charge ratio (m / z) of each product ion detected as a function of the precursor ion m / z values transmitted by the quadrupole mass filter. Since the precursor ion transmission window is scanned over time, the precursor ion m / z values transmitted by the quadrupole mass filter can also be thought of as times. The start and end times at which a particular product ion is detected are correlated to the start and end times at which its precursor is transmitted from the quadrupole. As a result, the start and end times of the product ion signals are used to determine the start and end times of their corresponding precursor ions.

[0007] Mass spectrum raw files generally contains all ion events in a form, such as cycle, time of flight (ToF) pulse, time bin, etc. This raw data, and DIA raw data files in particular, is often a prohibitively large file, limiting data processing and storage options.SUMMARY

[0008] Examples presented herein relate to a method of determining a confidence measure of precursor inference in mass spectrum data. The method includes obtaining mass spectrum data of a set of overlapping transmission windows, determining a response profile of the mass spectrum data across the set of overlapping transmission windows, evaluating the response profile using response profile metrics, and reporting the confidence measure of the precursor inference using the response profile and the evaluating of the response profile.

[0009] In other examples herein, the precursor inference is a predetermined precursor candidate. In further examples herein, the response profile metrics include an expected response profile. In yet other examples herein, the precursor inference is an unknown precursor identified based on the response profile.

[0010] In still other examples herein, the method further includes generating a response profile for a plurality of detected product ions in the mass spectrum data, generating a precursor inference for each product ion of the plurality of detected product ions, and grouping two or more product ions of the plurality of detected product ions using a common precursor inference. In further examples herein, the method further includes generating a compound fingerprint using the grouping.

[0011] In other examples herein, the response profile metrics include a symmetry of the response profile. In further examples herein, the method further includes identifying a slope of a rising linear function and a slope of a falling linear function of the response profile. In other further examples herein, the symmetry comprises a ratio of the slope ofthe rising linear function and the slope of the falling linear function. In still further examples herein, the ratio comprises an absolute value of each of the slope of the rising linear function and the slope of the falling linear function. In yet further examples herein, the response profile metrics further include tolerances for the symmetry determined using experimental data. In other further examples herein, the method further includes filtering one or more product ions associated with the response profile using the symmetry.

[0012] In still other examples herein, the response profile metrics include a quality measure determined by: calculating a rising fit for a rising linear function and a falling fit for a falling linear function of the response profile, and selecting one of the rising fit and the falling fit as the quality measure. In further examples herein, selecting one of the rising fit and the falling fit as the quality measure comprises selecting the lower of the rising fit and the falling fit. In other further examples herein, each of the rising fit and the falling fit are calculated as a coefficient of determination. In yet other further examples herein, each of fitting the rising linear function and the falling linear function comprises selecting a neighborhood of rising points and a neighborhood of falling points.

[0013] In still further examples herein, the method further includes identifying an apex of the response profile which bisects the neighborhood of rising points and the neighborhood of falling points. In yet still further examples herein, each of the neighborhood of rising points and the neighborhood of falling points includes at least two points. In other further examples herein, each of the neighborhood of rising points and the neighborhood of falling points includes at least three points. In yet further examples herein, selecting each of the neighborhood of rising points and the neighborhood of falling points comprises excluding the apex. In even further examples herein, excluding the apex further comprises excluding at least one additional point adjacent to the apex. In other still further examples herein, excluding the apex further comprises excluding at least one additional point adjacent to the apex to either side of the apex.

[0014] In other examples herein, the response profile metrics further include a symmetry of the response profile. In yet other examples herein, the method further includes calculating a precursor ion m / z using an intercept of the rising linear function and the falling linear function. In still other examples herein, the method further includes filtering one or more product ions associated with the response profile using the quality measure.

[0015] In still other examples herein, the encoding includes, for at least one unique product ion, a position of a window in which the at least one unique product ion appears and a sum of counts or intensities for a group of windows spanning the window in which the at least one unique product ion appears. In other examples herein, the method further includes calculating a precursor ion m / z for the precursor inference using an intercept of a rising linear function and a falling linear function of the response profile. In further examples herein, the method further includes filtering one or more product ions associated with the response profile using the precursor ion m / z to remove product ions attributed to a different precursor ion m / z. In yet still other examples herein, the encoding includes, for at least one unique product ion, a position of a window in which the at least one unique product ion appears and a sum of counts or intensities for a group of windows spanning the window in which the at least one unique product ion appears.

[0016] Other examples presented herein relate to a method of detecting conflicting relationships between precursors that are captured within the group of windows. The method includes obtaining an encoding of a scanning ion transmission window dimension of mass spectrum data, wherein the encoding includes, for a unique product ion, a position of a window in which the unique product ion appears and a sum of counts or intensities for a group of windows spanning the window in which the unique product ion appears, and filtering the sum at at least one position across the scanning ion transmission window dimension.

[0017] In other examples herein, filtering comprises running a subtraction filter. In further examples herein, the subtraction filter comprises subtracting an average intensity of neighboring positions defined by a neighborhood. In still further examples herein, the neighborhood comprises a number of positions away from the at least one position included in the average. In other further examples herein, the method further includes filtering the sum at each position across the scanning ion transmission window dimension.

[0018] In still other examples herein, filtering the sum at at least one position across the scanning ion transmission window dimension yields a filtered response profile, and the method further comprises filtering one or more product ions associated with the filtered response profile using a precursor ion m / z to remove product ions attributed to a different precursor ion m / z. In further examples herein, the precursor ion m / z is determined using the filtered response profile.

[0019] Other examples presented herein relate to another method of detecting conflicting relationships between precursors that are captured within the group of windows. The method includes obtaining an encoding of a scanning ion transmission window dimension of mass spectrum data, wherein the encoding includes, for at least one unique product ion, a position of a window in which the at least one unique product ion appears and a sum of counts or intensities for a group of windows spanning the window in which the at least one unique product ion appears, and generating, based on the encoding, a trace of the scanning ion transmission window dimension wherein each sum is substituted with a binary metric for presence of the at least one unique product ion in each window.

[0020] In other examples presented herein, the method further includes determining a precursor ion m / z corresponding to the at least one unique product ion as an m / z at which the at least one unique product ion is present in a highest total number of windows. In other examples presented herein, the encoding further includes a position and a sum for each of at least two unique product ions. In further examples presented herein, the trace includes each of the at least two unique product ions.

[0021] Other examples presented herein relate to a method of identifying a precursor m / z. The method includes obtaining an encoding of mass spectrum data, fitting a rising linear function and a falling linear function to a response profile of the mass spectrum data, determining an intercept between the rising linear function and the falling linear function, and identifying the precursor m / z using the intercept.

[0022] In other examples presented herein, the apex is excluded from at least one of the rising linear function and the falling linear function. In further examples presented herein, the apex is excluded from each of the rising linear function and the falling linear function. In other examples herein, each of the rising linear function and the falling linear function is calculated using at least two points. In still other examples presented herein, each of the rising linear function and the falling linear function is calculated using at least three points. In further examples presented herein, each of the at least three points is three consecutive points.

[0023] In yet other examples presented herein, the method further includes determining a symmetry of the response profile and generating a confidence measure of the response profile for precursor inference using the symmetry. In further examples presented herein, the symmetry comprises a ratio of absolutes values of a slope of the rising linear function and a slope of the falling linear function.

[0024] In still other examples presented herein, the method further includes determining a quality measure by: calculating a rising fit for the rising linear function and a falling fit for the falling linear function, and selecting a lower of the rising fit and the falling fit. In yet other examples presented herein, the encoding is of the scanning ion transmission window dimension includes, for at least one unique product ion, a position of a window in which the at least one unique product ion appears and a sum of counts or intensities for a group of windows spanning the window in which the at least one unique product ion appears.

[0025] A variety of additional inventive aspects will be set forth in the description that follows. The inventive aspects can relate to individual features and to combinations of features. It is to be understood that both the forgoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the broad inventive concepts upon which the embodiments disclosed herein are based.BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The accompanying drawings, which are incorporated in and constitute a part of the description, illustrate several aspects of the present disclosure. A brief description of the drawings is as follows:

[0027] FIG. 1 A is an example QI trace before filtering according to embodiments of the present disclosure.

[0028] FIG. IB is an example QI trace following filtering according to embodiments of the present disclosure.

[0029] FIG. 2 is a block diagram of an example mass spectroscopy system.

[0030] FIG. 3 is block diagram of an example data processing system for processing and evaluating mass spectrum data.

[0031] FIG. 4 is a flowchart of a method of determining a confidence measure of precursor inference in mass spectrum data.

[0032] FIG. 5A is an example of a response profile, depicted as a trace, with low symmetry.

[0033] FIG. 5B is an example response profile, also depicted as a trace, with high symmetry.

[0034] FIG. 6 is an example response profile including both a rising linear regression and a falling linear regression.

[0035] FIG. 7 is a flowchart of an example method of detecting conflicting relationships between precursors that are captured within a group of windows.

[0036] FIG. 8 is an example response profile depicting results of the method of FIG. 7.

[0037] FIG. 9 is a flowchart of another example method of detecting conflicting relationships between precursors that are captured within a group of windows.

[0038] FIG. 10 is an example response profile depicting results of the method of FIG. 9.

[0039] FIG. 11 is a flowchart of a method of identifying a precursor m / z.

[0040] FIG. 12 illustrates an example block diagram of a virtual or physical computing system.DETAILED DESCRIPTION

[0041] Reference will now be made in detail to exemplary aspects of the present disclosure that are illustrated in the accompanying drawings. Wherever possible, the same reference numbers will be used throughout the drawings to refer to the same or like parts.

[0042] Disclosed herein are systems and methods for improved processing of mass spectrum data. Existing mass spectrum data processing methods, and particularly those methods for data independent acquisition (DIA), are often limited to being executed on the data as it is acquired, requiring large amounts of processing power and permitting only a limited time window in which data can be analyzed. Some existing solutions provide for compression of the data to allow for storage, but the resulting loss of some data points due to the compression leads to degradation of analytical results. Disclosed herein are systems and methods for improved processing and analysis of compressed mass spectrum data and, in some embodiments, particularly DIA compressed data. The systems and methods disclosed herein provide for improved reliability and utility of the compressed data.

[0043] One example of compressing and storing mass spectrum data is given in International Publication No. WO 2020 / 240506 Al (“the ‘506 Publication”), which describes the conversion of a raw DIA file to a particular file (e.g., a particular wiff file format) for processing and storage. In the ‘506 Publication, QI information is encoded in large bins, such as 1 / 5 of the QI filter window (Da). The entirety of the ‘506 Publication is incorporated herein by reference.

[0044] While this binning of the data provides effective compression and furthers the practical consideration of storage, some data points are inevitably lost during the compression process. Fragment precursor specificity may be improved by encoding with less binning or smaller bins, for example by using a smallest bin size determined based on the setting of a precursor ion filter (e.g., a first quadrupole filter size). However, such a change in the bin size would substantially increase the file size and therefore does not provide a practical alternative for storage considerations.

[0045] Compressed encoding of mass spectrum data also introduces a convolution in reducing the file size for storage. For example, for DIA data, the QI dimension becomes blurred, making it difficult to distinguish co-isolated precursors from an encoded QI trace of a shared fragment. In qualitative analysis, accurate precursorfragment relationships are important for comprehensive compound identification. In addition, intensity portions of a shared fragment with respect to other unique fragments of a given precursor need to be accurate for effective analysis to be performed. Thus, following compression of a mass spectrum file, a problem arises in how to extract all fragments, both unique and shared, for a given precursor, from the data in a way that each fragment’s relative signal is representative of the true amount.

[0046] One possible solution to this problem was proposed in International Publication No. WO 2023 / 017448 Al (“the ‘448 Publication”), which is incorporated herein by reference in its entirety. Instead of encoding in equal size bins, raw data may be encoded in a new file format having experiment width as a function of fragment m / z. This example format provides high fragment-precursor specificity for each observed ion. The specificity degrades with ion intensity, e.g., lower intensity ions will have large precursor uncertainty. Maximum specificity would be achieved for ions with an ion rate per time of flight (ToF) pulse close to 1. These ions would have precursor specificity equal to QI filter step value in Da. For ions that have only a single ion event during QI scanning time over the corresponding precursor, specificity may be as low as precursor isolation filter window size itself. However, it is noted this process is not limited to ToF pulse rate, but also allows for the application of one or more of binning an encoding.

[0047] However, this method is not practical with current data file formats and software tools. Variants of this method could be used for data deconvolution after binning and encoding, but the time and computation power consumed are prohibitive. While the ‘448 Publication provides high qualitative and quantitative accuracies for real-time mass spectra, its principles are not readily applied to data compressed for storage , like thatexemplified by the ‘506 Publication, in order to offer compatibility with existing data processing workflows. In order to achieve quality analysis of compressed mass spectra data, such as may be retrieved from storage some time after a sample is run, systems and methods for reconstructing the data to account for what is lost during compression are needed. Some existing methods for reconstructing data include deblurring and nonnegative least squares.

[0048] Disclosed herein are systems and methods for evaluating, clarifying, and preforming precursor inference on raw or compressed, or otherwise preprocessed, mass spectrum data, among other concepts and principles. In some embodiments, the specificity of the encoded data for a further qualitative analysis is improved by generating a trace of the data. For example, an alternative trace, such as a binary trace, may be regenerated from QI encoded traces, allowing for both qualitative as well as accurate quantitative decompositions. This data transformation can be applied to, for example, an existing wiff file, improving the specificity of the information in the file. The precursor m / z can be determined with an improved accuracy (e.g., 0.1 Da accuracy in some example use cases using a 5 Da isolation window width). Accuracy may be increased by using narrower isolation windows.

[0049] In some embodiments, relationships among fragments otherwise obscured by typical intensity traces are revealed by the systems and methods disclosed herein. In this way, unrelated fragments can be identified and filtered. As will be discussed in greater detail below, for a given cycle, a QI trace may be sharpened, e.g., made more selective by the application of the methods and principles of the present disclosure.

[0050] Referring now comparatively to FIG. 1A and FIG. IB, an example QI trace is shown prior to applying (FIG. 1 A) and following applying a QI filter across all cycles, according to embodiments of the present disclosure (FIG. IB). The example presented is FIGS. 1A and IB depicts the disclosed filtering in method in combination with LC deconvolution, but those skilled in the art will recognize that the QI filtering may also be applied on a single cycle or otherwise independent of an LC deconvolution. As shown in FIG. 1A, a QI trace may appear as a linear combination of “noisy triangles.” A well- defined triangle is generated in response to high intensity fragments which are not interfered with, an irregular blobby” triangle is generated in response to low intensity and / or interfered (shared) fragments, and a trapezoid or single rectangle trace is generated in response to very low intensity fragments (e.g., one or only few ion events during dwelltime). In these traces, high intensity fragments may obscure relevant data within the trace generated in response to lower intensity fragments.

[0051] As disclosed in some embodiments herein, a subtraction filter is run for all m / z-bins across a QI or experimental dimension at each LC cycle. The “subtraction” preserves the signal for flat-tops for low intensity fragments, while sharpening high intensity fragments, as can be seen in comparing FIGS. 1A and IB. In this way, the disclosed subtraction filter provides at least two advantages. First is the relative enhancement of fragments is spectrums which may include fragments from two or more QI windows, with fragments relevant to a target narrow precursor range being “shadowed” or otherwise obscured by more intense interferences from a wider range. Second is detection of interferences, where after the filter is run it is possible to detect two or more QI trace apexes separated by less than the QI width, whereas before the filter is run the trace may appear to have only one broad apex on an irregular (e.g., “blobby”) trace.

[0052] The sharpening of the trace visible in FIG. IB relative to FIG. 1A is a result of intensity subtraction of, for example, an average intensity of neighboring QI points or experimental data points. The sharpening is best exemplified by traces 10a and 30a, which dominate in FIG. 1A and are reduced to traces 10b and 30b, each with a narrower base and a sharper peak in FIG. IB. In the sharpening of traces 10a and 30a, the contribution of another fragment, trace 20a, becomes visible in FIG. IB as trace 20b, while being almost fully obscured by traces 10a and 30a in FIG. 1A. The filtering described herein reduces or removes intensity contributions from close-by precursors. The resulting spectrum (FIG. IB) has less interference signal. In some embodiments, all fragments represented by triangles with different apexes in QI dimension are subtracted (e.g., eliminated or suppressed) at a current QI filtering point or experiment.

[0053] Benefits of the filtering include the visible sharpening of the QI trace. When subtraction is applied to an analog intensity traces, which may deviate from a theoretical triangle shape, significant or complete intensity reduction for unrelated fragments may be achieved. As discussed herein, unrelated fragments are fragments originating from a precursor ion that is more than an encoding window apart from a target encoding window center.

[0054] In embodiments, the intensities after subtraction are affected by interfering precursor contribution amount and may not be reliable for relative quantitation. After the subtraction, the resulting spectra retains fragments unique for a narrower precursor range,e.g., about 2x narrower, than an originally encoded spectra. Enhancing specificity of the DIA or other MS data is especially beneficial in workflows that do not include an orthogonal separation (e.g., by LC, DMS, etc.) The solution presented herein preserves intensity profiles across orthogonal separation, making it useful preprocessing step or, in embodiments, analysis according to methods of the present disclosure may be combined with orthogonal data dimensions (e.g., LC, mobility, etc.). The filtered QI trace may be used to infer a precursor m / z and / or filter an MSMS spectra to remove fragments not resembling, and therefore likely not being related to, the particular precursor m / z. The precursor m / z may be determined, for example, using an MSI spectra (an untargeted method) or from a reference library (a targeted method).

[0055] Methods presented herein, such as the subtraction method discussed above, retain an intensity fraction of shared fragments that is representative of relative fragment contribution to a precursor of interest or, otherwise stated, the relative fragment ratios are better represented in the deconvoluted spectrum. This is especially true for fragments shared by precursors separated by a distance of approximately twice a step of the encoding windows in Da, or by about 30% or more of the scanning QI window width (e.g., for a 5:1 encoding scheme).

[0056] A ratio of the “QI subtracted” intensity to the original intensity is indicative of precursor-to-fragment “exclusivity,” or whether it is shared by a co-isolated precursor. This feature may be useful when selecting quantifier fragments, as well as in determining an uncertainty of fragment- precursor relationships useful in library searches as a fragment weighting factor.

[0057] In embodiments, systems and method are further disclosed herein for evaluating an encoding of mass spectrum data to determine a confidence measure of the suitability of the data for producing a precursor inference. As discussed herein, the confidence measure provides a value indicative of the suitability or reliability of a particular response profile to provide a reliable inference of the precursor ion giving rise to a particular fragment or product ion.

[0058] In some embodiments, a symmetry of a QI trace is calculated as a confidence measure, for example as a ratio of absolute values of slopes of rising and falling lines. In some embodiments, the fit of both a rising and a falling function are calculated as the confidence measure, for example as a coefficient of determination (e.g., R-squared value). In embodiments, the rising and falling functions are generally linear functions, and are so discussed herein. However, those of skill in the art will recognize that in somecases the functions will not be linear, such as in cases implementing narrower isolation windows. In examples, the lower of the two fits is considered as a measure of the quality of the QI trace for use to determine a precursor inference. The value of the QI quality measure indicates a sufficient signal (e.g., above noise) for a precursor m / z inference determination

[0059] In some embodiments, a particular precursor inference candidate is sought in a targeted manner. In such embodiments, the value of the QI quality measure indicates the presence of precursors close to the candidate that share the fragment of interest. In some embodiments, candidate fragments belonging to a specific precursor are filtered based on one or more of a calculated precursor m / z, QI symmetry, and QI quality. The response profile evaluation and filtering described here may further be optionally combined with further analytical methods, such as when applied to LC / MS data or otherwise combined with orthogonal data dimensions (e.g., a time dimension, a mobility dimension, etc.). Some non-limiting examples of further analytical methods which may be used in combination with the methods disclosed herein include principal component analysis (PCA), MultiVariate analysis, pattern recognition or other machine learning analysis, and principal component variable grouping (PCVG).

[0060] In embodiments, systems and methods for performing precursor inference are also present herein. Encoded data is typically down-sampled which reduces precursor inference accuracy. To overcome this precursor inference accuracy loss, a “QI triangle” centroid may used to determine the precursor m / z. However, this approach is sensitive to noise and does not consider the contribution of isotope precursors into the monoisotopic fragment signal. Disclosed herein is an improved means of performing precursor inference by performing linear regression on the experimental data constituting the sides of peaks in an encoded QI trace to reconstruct the triangles for a fragment. The intercept of the lines on rising and falling parts of the peak represents a precursor m / z, which may be conventional monoisotopic m / z or a weighted average m / z of an isotopic cluster with the weight determined according to a portion of the cluster isolated within the isolation window. In embodiments, the precursor ion m / z calculated using the intercept is not a conventional monoisotopic m / z. For example, if an ion is multiply charged or the isotopic distribution is not a best fit for a monoisotopic m / z, the intercept may instead indicate an isotope cluster. The isotope cluster may be defined with a weighted average m / z and therefore be shifted to the right by an amount dependent on a portion of the isotope pattern isolated by the QI isolation window. In examples, this approach is able todetermine the precursor m / z with accuracy commensurate to about one tenth of the compression size (e.g., mass spectrum data using a 5Da QI window and encoded by binning every five steps of the QI window may determine the precursor m / z with accuracy of 0.1 Da). In embodiments, a fragment may be identified as having a relationship with two or more co-isolating precursor ions. For example, based on a peak found to represent an isotopic cluster.

[0061] FIG. 2 is a block diagram of an example mass spectrometry system 100. In embodiments, other systems for determining compound structure and identity may be integrated with a mass spectrometry system or, in other embodiments, may be independent of the mass spectrometry system. Example system 100 includes an ion source 110, a first mass separator 120, a fragmentation device 130, a second mass separator or a mass analyzer 140, and a computing system 150.

[0062] In embodiments, system 100 further includes a sample introduction device 160. Sample introduction device 160 introduces one or more compounds of interest from a sample to ion source 110 over time. Sample introduction device 160 performs techniques that include, but are not limited to, direct injection, liquid chromatography, gas chromatography, capillary electrophoresis, or ion mobility.

[0063] Mass filter 120 and fragmentation device 130 are shown as different stages of a quadrupole and mass analyzer 140 is shown as a time-of-flight (TOF) device. Those of ordinary skill in the art will appreciate that either of mass filter 120 and mass analyzer 140 may include other types of mass separator and analysis devices including, but not limited to, ion traps, orbitraps, ion mobility devices, time-of-flight (TOF) devices, or Fourier transform ion cyclotron resonance (FT-ICR) devices. In embodiments, mass filter 120 and mass analyzer 140 are respective examples of a first and a second mass separator, arranged in a series. A system may be configured according to the present disclosure with a quadrupole for the first mass separator, or mass filter 120, and a TOF device for the second mass separator, or mass analyzer 140. Each mass separator is configured to receive a set of ions, perform a detection of the set of ions, and generate a set of detection signals corresponding to detection of the set of ions.

[0064] Ion source device 110 transforms a sample or compounds of interest from a sample into an ion beam. Ion source device 110 can perform ionization techniques that include, but are not limited to, matrix assisted laser desorption / ionization (MALDI) or electrospray ionization (ESI).

[0065] Mass filter 120 receives the ion beam. In embodiments, mass filter 120 is configured by a user for a particular precursor ion transmission window based on the experimental goals for the sample being run. The precursor ion transmission window, as discussed herein, refers to the range of precursor or parent ions that are allowed to pass through a specific selection step and into the subsequent stages of mass analysis or fragmentation. In many tandem mass spectrometry (MS / MS) experiments, the precursor ions are first filtered based on their m / z (mass-to-charge ratio) in order to isolate a specific ion or group of ions of interest for further analysis or fragmentation. The precursor ion selection process employs a mass filter or a specific set of voltages that allow only ions within a certain m / z range (the precursor ion transmission window) to pass through to the next stage.

[0066] Fragmentation device 130 of tandem mass spectrometer 102 fragments or transmits the precursor ions transmitted at each overlapping step by mass filter 120. In examples related to scanning DIA, one or more resulting product ions are produced for each overlapping window of the series. Fragmentation device 130 fragments the precursor ions when a collision energy high enough to fragment ions is used. Fragmentation device 130 transmits the precursor ions when a collision energy low enough not to fragment ions is used. As a result, the resulting product ions can include precursor ions.

[0067] In mass spectrometry, dissociation mechanisms are techniques used to fragment molecules into smaller pieces to facilitate their analysis and identification. For example, collision energy is used to influence the fragmentation of ions during collision- induced dissociation (CID) or collision-induced fragmentation (CIF). This process is commonly used in MS / MS to provide structural information about a molecule by inducing the dissociation of parent or precursor ions into fragment ions. The setting of the collision energy impacts the resulting fragmentation pattern and may influence how meaningful any acquired data is. Some other examples of dissociation mechanisms include electron transfer dissociation, electrospray ionization, matrix-assisted laser desorption / ionization, infrared multiphoton dissociation, and sustained off-resonance irradiation. Each dissociation method offers varying advantages depending on the type of analysis being performed and the nature of the sample. The choice of dissociation technique may impact the type of information obtained from mass spectrometric analysis.

[0068] Mass analyzer 140 of tandem mass spectrometer 102 detects intensities or counts for each of the one or more resulting product ions for each overlapping window of the series that form mass spectrum data for each overlapping window of the series.

[0069] Computing system 150 can be, but is not limited to, a computer, a microprocessor, the computing system of FIG. 12, or any device capable of sending and receiving control signals and data from a tandem mass spectrometer and processing data. Computing system 150 is in communication with ion source device 110, mass filter 120, fragmentation device 130, and mass analyzer 140. Computing system 150 is shown as a separate device but can be a processor or controller of tandem mass spectrometer 102 or another device. Computing system 150 may store in a memory device (not shown) mass spectrum data for each precursor ion window analysis is performed for, including for each overlapping window of the series in examples performing scanning DIA.

[0070] In embodiments, computing system 150 instead performs an encoding and storing step, and encodes and stores each unique product ion detected by mass analyzer 140 in real-time during data acquisition. Prior to storing mass spectrum data, computing system 150 performs one or more processing steps on the raw mass spectrum data received to prepare the data for viewing, analysis, and storage. Raw mass spectrum data includes the counts or intensities of product ions at different m / z ratios over time. Computing system 150 may be further configured to retrieve stored mass spectrum data for subsequent, additional processing and analysis, and / or transmission to additional, remote or otherwise separate computing systems.

[0071] FIG. 3 is block diagram of an example data processing system 200 for processing and evaluating mass spectrum data, including encoded or compressed mass spectrum data. Data processing system 200 is implemented, in embodiments, by computing system 150. Data processing system 200 may be among multiple subsystems or software executed by computing system 150 in operating mass spectrometer 102 and handling of data output by the mass spectrometer. The present disclosure is directed to data processing for evaluation and analysis, but those of skill in the art will readily understand that other subsystems and / or modules may exist within and be executed by computing system 150. Computing system 150 is presented in examples herein as a single device, but in embodiments may be one or more processing devices networked or otherwise in communication. Functions may be divided among individual devices or shared across the collective processing capability of the one or more processing devices. In embodiments, data processing system 200 forms part of or serves as a controller. Thecontroller may be configured to transmit operation commands to the mass spectrometer and / or to receive unprocessed mass data from second mass separator including the set of detection signals and perform one more data processing actions on the unprocessed mass data. In embodiments, data processing system 200 may be fully remote from and independent of the mass spectrometry system 100.

[0072] In the example of FIG. 3, data processing system 200 includes storage 204, response profiler 210, confidence analyzer 220, deconvolution module 230, precursor inference 240, and display 250. Data, including mass spectrometry data 202 and derivative processing outputs of mass spectrometry data 202, may be passed freely among the components of data processing system 200. Various outputs of the analysis and processing components may be stored in storage 204, or pass to display 250.

[0073] Storage 204 receives mass spectrometry data 202 from mass spectrometer 102 for storage. In embodiments, mass spectrometry data 202 may instead be received from a computing system, such as another computing system 150, and undergo processing and / or preprocessing before being received by storage 204. Mass spectrometry data 202 may be raw or processed and encoded, including being compressed, prior to being received and stored at storage 204.

[0074] Raw MS data generally consists of the unprocessed measurements collected directly from the instrument and is often stored in vendor-specific binary formats, or converted into open formats like mzML or mzXML. This data comprises sequential spectra, each containing lists of mass-to-charge (m / z) values and associated intensity measurements, often with orthogonal data dimension, such as retention times, and metadata. The structure of raw data is irregular because the m / z spacing depends on instrument resolution, and the files generally include noise, baseline fluctuations, and other acquisition artifacts. Due to its high dimensionality and variability, raw MS data is generally not immediately suitable for direct comparison or computational analysis by conventional methods.

[0075] Preprocessing is applied to improve data quality, consistency, and interpretability. Common steps include noise reduction and baseline correction to minimize low-intensity fluctuations, peak detection or centroiding to convert continuous profile spectra into discrete m / z-intensity peaks, and spectral alignment to correct for small shifts in m / z or retention time between runs. Normalization techniques may be used to adjust for differences in total signal intensity, and feature extraction methods group related peaks into higher-level structures such as isotopic envelopes or chromatographicfeatures. At this stage, the data is cleaned and organized but remains unencoded, meaning that it still represents measured values directly, often in peak tables or processed mzML files.

[0076] Compression steps, such as binning and encoding, are optional but frequently employed steps in modem workflows to facilitate downstream analysis. Binning involves partitioning the m / z range into fixed-width intervals and aggregating the intensities within each bin. This produces uniformly spaced data, improving comparability across samples and instruments and reducing dimensionality. Encoding transforms the preprocessed, and often binned, data into compact, standardized representations. Conventional approaches include sparse encodings, which store only nonzero intensities; vector encodings, which represent spectra as fixed-length arrays; and compression into storage formats. Encoding is particularly useful for machine learning and high- throughput comparative analyses, where uniform data structures are required.

[0077] Response profiler 210 may receive mass spectrometry data 202 from storage 204, or from mass spectrometer 102 or another computing system, and generate a response profile. The response profile provides the response of a particular fragment or product ion of interest at the stored data points. The response profile may be generated, in examples, as a trace, as a data table, or another organization of the relevant data.

[0078] Confidence analyzer 220 evaluates a response profile for relative quality to be used for performing a precursor inference. Confidence analyzer 220 may have one or more modules for different confidence measures or response profile metrics. For example, a symmetry analyzer 222 performs a symmetry analysis on a response profile to determine a confidence measure. Quality measure 224 may perform a line fitting analysis on a response profile to determine a confidence measure.

[0079] Deconvolution module 230 performs processing on a response profile to detect or reduce interferences from isotopic alternate precursor ions. Deconvolution module 230 may have one or more modules to perform different deconvolution methods such as, subtraction filter 232 and binary trace converter 234.

[0080] Precursor inferencer 240 performs precursor inference analysis on a response profile. The precursor inference analysis may be performed in either a targeted manner, by targeted module 242, or a survey manner, by untargeted module 244.

[0081] FIG. 4 is a flowchart of a method 300 of determining a confidence measure of precursor inference in mass spectrum data. In embodiments, the method 300 is performed by a data processing system, such as data processing system 200 of FIG. 3.As discussed herein, a confidence measures refers to a value indicating a relative reliability of a particular set of mass spectrum data for providing an inference of the precursor ion giving rise to a particular product ion. In embodiments, a precursor inference may be employed in a targeted manner and directed at a particular predetermined candidate precursor ion. The predetermined candidate precursor ion may be selected from a library of ions, or may be determine based on prior analysis or knowledge. For example, previous experimental or in silico data, such as a preliminary scan, e.g., a low energy scan, performed on the same sample or a sample of the same source.

[0082] In other embodiments, the precursor inference may be employed in an untargeted or survey manner, and a precursor inference determined from the data without having a predetermined precursor ion candidate. In examples, the manner of precursor inference applied may influence the choice and application of confidence measure analysis.

[0083] At operation 302, mass spectrum data of a set of overlapping transmission windows is obtained. In embodiments, the mass spectrum data is an encoding of data which has been stored or processed (e.g., compressed) between being acquired by the mass spectrometer and being obtained at operation 302. In embodiments, the encoding includes, for at least one unique product ion, a position of a window in which the at least one unique product ion appears and a sum of counts or intensities for a group of windows spanning the window in which the at least one unique product ion appears.

[0084] At operation 304, a response profile of the mass spectrum data across the set of overlapping transmission windows is generated. As discussed herein, a response profile refers to a record of a particular fragment’s or product ion’s response at one or more acquisition window positions. In examples, a response profile includes a product ion’s response at each acquisition window position of a scanning sequence. In some embodiments, the response profile is a trace, such as a trace of fragment ion intensity mapped across each acquisition window.

[0085] At operation 306, the response profile is evaluated using response profile metrics. Response profile metrics refer to one or more calculations performed using the response profile or reference data to which the response profile or calculation is compared. At operation 308, the confidence measure can be reported using the evaluation of the response profile. The confidence measure may be reported as a value of theresponse profile metric or a notification of satisfactory profile quality (e.g., a satisfactory or unsatisfactory grade, a color / letter / number grade on a graduated scale, etc.).

[0086] In some embodiments, the response profile metrics include a symmetry of the response profile. Determining the symmetry of the response profile may include identifying a slope of a rising linear function and a slope of a falling linear function of the response profile. The symmetry may be determined as a ratio of the slope of the rising linear function and the slope of the falling linear function. For example, a ratio of an absolute value of each of the slope of the rising linear function and the slope of the falling linear function may be used to determine the symmetry. Theoretical QI traces for all fragments will generally have a symmetry of 1. In some examples, the response profile metrics may include a lower threshold, e.g., 0.9, for the symmetry. In some examples, the response profile metrics may further include an upper threshold, e.g., 1.1, for the symmetry. In embodiments, the response profile metrics include tolerances for the symmetry determined using, for example, experimental data or chemical knowledge. Chemical knowledge may be applied using a known isotope pattern distribution in associated with the particular QI window used. For example, a particular isotope pattern with result in different trace symmetry depending on the characteristics of the QI window. Likewise, a particular QI window can produce different trace symmetry based on variations in ion charge and isotope pattern.

[0087] In embodiments, the symmetry is determined using a statistical analysis, e.g., skewness.

[0088] In embodiments, one or more product ions associated with a candidate precursor ion may be filtered within the response profile using the symmetry. For an example use case of a singly charged molecule with an encoding step that corresponds to the spacing of isotope peaks (IDa), an intercept of the QI trace will be at a precursor m / z for a remaining precursor in the MS / MS data. For smaller fragments, the intercept will provide an estimate that can be further enhanced by considering the contribution of one or more precursor isotopes to fragment’s monoisotopic peak. While overall bias may vary with the elemental composition of the precursor and the fragments, in general, the smaller the fragment, the bigger the bias in the precursor estimate will be.

[0089] Points at the top and bottom of the QI trace may be affected by the accurate position of the precursor in its QI encoding step boundaries. An extreme case is the precursor positioned at the boundary of two difference QI encoding steps, where the QI trace will have a trapezoid shape. The limits and confidences for the assessment of thesymmetry for the experimental QI traces within the data processing may be established by determining the symmetries for the remaining precursors, particularly within the mass range of interest, under the same acquisition and encoding conditions as the experimental data.

[0090] The symmetry analysis can thus provide useful information in assessing the response profile for the contribution of fragments from more than one precursor ion (e.g., one or more isotopic precursor ions) and enable more effective filtering to generate a response profile reliably associated with a single precursor ion.

[0091] FIGS. 5A and 5B are a comparative examples of response profiles. FIG. 5A is an example of a response profile, depicted as a trace, with low symmetry, indicating low strength of a signal of interest and potential interference by one or more additional precursor ions. In contrast, FIG. 5B is an example response profile, also depicted as a trace, with high symmetry, indicating a good quality trace for providing a high confidence inference of a single mono-isotopic precursor m / z. While the examples of FIGS. 5 A and 5B are depicted as traces, it will be understood by those of skill in the art that symmetry may also be effectively evaluated in data otherwise presented, e.g., a data table.

[0092] In some embodiments, the response profile metrics include a quality measure. The value calculated of the quality measure may be used to indicate a sufficient signal (above noise) for a precursor m / z inference to made with confidence. The value may further indicate a presence of precursors close to a candidate precursor m / z that shares the fragment of interest.

[0093] The quality measure may be calculated using a rising fit for a rising linear function and a falling fit for a falling linear function of the response profile and selecting one of the rising fit and the falling fit as the quality measure. For example, the lower of the rising fit and the falling fit may be selected as the quality measure. In some embodiments, one or each of the rising fit and the falling fit are calculated as a coefficient of determination or R-squared value. In embodiments, the rising and falling functions are generally linear functions, and are so discussed herein. However, those of skill in the art will recognize that in some cases the functions will not be linear, such as in cases implementing narrower isolation windows.

[0094] Each of the rising linear function and the falling linear function may be fit by selecting a neighborhood of points, such as a neighborhood of rising points and a neighborhood of falling points. In embodiments, the response profile includes an apexwhich bisects the neighborhood of rising points and the neighborhood of falling points and may be identified as part of calculating or evaluating the response profile. In some examples, one or each of the neighborhood of rising points and the neighborhood of falling points includes at least two points, at least three points, etc. For example, in some preferred embodiments at least three points for each neighborhood may be preferred, with a fit based on at least two points used where three points are not available. For example, at least three points may be desired for each of the rising and the falling fit, but only two points available for either the rising fit or the falling fit. In such a case, one of the rising fit and falling fit is calculated using the preferred three points and one is calculated using the available two points.

[0095] In some embodiments, the apex is excluded from one or both of the neighborhood of rising points and the neighborhood of falling points. In some examples, excluding the apex further also excludes at least one additional point adjacent to the apex. Excluding the apex may also exclude at least one additional point adjacent to the apex to either side of the apex.

[0096] FIG. 6 is an example response profile 350 including both a rising linear regression 352 and a falling linear regression 354. In the example response profile 350, apex 356 is excluded from both linear regressions. Rising linear regression 352 includes a neighborhood of three points rising to the apex 356 and falling linear regression includes a neighborhood of points falling away from the apex 356.

[0097] In embodiments, one or more product ions associated with the response profile are filtered using the quality measure. In some embodiments, the quality measure and the symmetry may be used together as response profile metrics to evaluate one response profile or a set of response profiles. In some further examples, a precursor ion m / z inference is determined using an intercept of the rising linear function and the falling linear function. In embodiments, one or more product ions associated with the response profile are filtered using the precursor ion m / z. In some embodiments, the method 300 optionally proceeds further to calculating a precursor ion m / z for the precursor inference using an intercept of a rising linear function and a falling linear function of the response profile. In some examples, the method further optionally includes filtering one or more product ions associated with the response profile using the precursor ion m / z, e.g., without also filtering for one or either of the symmetry or quality measure. In embodiments, filtering the response profile for the precursor ion m / z may provide anindication of whether the fragment has a relationship with two or more co-isolating precursor ions.

[0098] In targeted embodiments, where a particular candidate precursor ion m / z is sought out, the response profile metrics may include an expected response profile. In survey embodiments, which are implemented in an untargeted manner and where there is not a particular candidate precursor ion m / z being sought, response profile metrics may instead focus on the precision with which a “possible” precursor m / z inference may be reached.

[0099] At operation 308, a confidence measure of the precursor inference is reported using the response profile and the evaluating performed at operation 306. As discussed herein, reporting may refer to displaying a calculated value for the confidence measure, or displaying the values calculated from one or more response profile metrics, e.g., a symmetry value or a quality measure value.

[0100] In some embodiments, a response profile is generated for each of a plurality of detected product ions in the mass spectrum data. A precursor inference may be generated for each product ion of the plurality of detected product ions. Two or more product ions of the plurality of detected product ions may then be grouped using a common precursor inference. The grouping may be used to generate a compound fingerprint, such as by combining several groupings from a compound that gave rise to multiple precursors. In embodiments for untargeted data analysis, the grouping is used to calculate a representative intensity profile for the group and each product ion within the group is scored with respect to representative intensity profile. In embodiments, the response profile is calculated as a matrix values for each product ion and grouping the product ions is based on product ions having a matrix value within a predetermined tolerance.

[0101] FIG. 7 is a flowchart of a method 400 of detecting conflicting relationships between precursors that are captured within a group of windows. In embodiments, the method 400 is performed by a data processing system, such as data processing system 200 of FIG. 3.

[0102] At operation 402, an encoding of a scanning ion transmission window dimension of mass spectrum data is obtained. In embodiments, the encoding includes, for a unique product ion, a position of a window in which the unique product ion appears and a sum of counts or intensities for a group of windows spanning the window in which the unique product ion appears.

[0103] At operation 404, at least one data point from the encoding is filtered. For example, in embodiments where the encoding includes a sum and a position for each product ion, operation 404 includes filtering the sum at at least one position across the scanning ion transmission window dimension. In embodiments, the filtering is a subtraction filter. In some embodiments, the subtraction filter includes subtracting an average intensity of neighboring positions defined by a neighborhood. A neighborhood may be defined as a number of positions away from the at least one position included in the average.

[0104] In examples, the filter may use I* values at a given retention time to filter out intensities with an HI* threshold, where I* is corrected intensity I of a given experiment e of the encoded file. I* is calculated as:Where de defines a neighbourhood around experiment e to be considered when creating a line-segment. Quantitative analysis uses original / intensities but corrected / * intensities provide useful qualitative results to improve deconvolution and remove interferences.

[0105] At operation 406, a determination is made whether additional points remain to be filtered. In embodiments, filtering is performed at each point across a range of points. The range of points may include a full dimension of the points in the data, for example each point across a QI or experimental dimension of the data may be filtered. In embodiments where the encoding includes a sum and a position for each product ion, the sum at each position across the scanning ion transmission window dimension may be filtered. In such cases, the operation 406 continues to return to operation 404 to filter additional points until all points across the experimental dimension have been filtered.

[0106] Once filtering is complete, the filtered data is stored and / or presented, at operation 408. For example, the filtered data may be stored in an integrated or separate non-transitory memory, for subsequent retrieval and viewing and / or further processing. The filtered data may be presented directly and modified or otherwise further processed based on user input. In embodiments, the filtered data may be stored before additional processing and after additional processing for comparison and other analysis.

[0107] FIG. 8 is an example response profile 450 depicting results of the method 400. An original response profile 452 can be seen to be filtering into a sharper response profile 454 following the execution of the method 400. Original data, which may becompressed data retrieved from storage, used to generate response profile 452 is shown below in Table 1:Table 1.data for the fragment with m / z 204 resulting from the execution of the method 400:Table 2.Following subtraction, the indicated precursor m / z is 342.13, which is more accurate to the target precursor m / z of 342.142.

[0108] FIG. 9 is a flowchart of a method 500 of detecting conflicting relationships between precursors that are captured within a group of windows. In embodiments, the method 500 is performed by a data processing system, such as data processing system 200 of FIG. 3.

[0109] At operation 502, an encoding of a scanning ion transmission window dimension of mass spectrum data is obtained. In embodiments, the encoding includes, for a unique product ion, a position of a window in which the unique product ion appears and a sum of counts or intensities for a group of windows spanning the window in which the unique product ion appears.

[0110] At operation 504, a trace of the scanning ion transmission window dimension is generated, based on the encoding. At operation 506, the intensity value of the trace is substituted with a binary metric for presence of the fragment at each point. For example, in embodiments where the encoding includes a sum and a position for each product ion, each sum is substituted with a binary metric (e.g., 1 for presence and 0 for absence) for presence of the at least one product ion in each window. The binary trace may be used for detection of shared fragments, or fragments having a relationship with two or more co-isolating precursor ions.

[0111] In some embodiments, the method 500 may optionally extend to operation 508 and a precursor ion m / z corresponding to the at least one unique product ion is determined as an m / z at which the at least one unique product ion is present in a highest total number of windows overlapping with the corresponding point. In some embodiments, such as embodiments implemented using an encoding including position and intensity data for each of at least two unique product ions, the trace includes each of the at least two unique product ions.

[0112] FIG. 10 is an example response profile 550 depicting results of the method 500. Trace 552 is an overall intensity trace. Trace 554 is a trace resulting from method 500 showing contributions from a first precursor. Trace 556 is a trace resulting from method 500 showing contributions from a second precursor. In result, the contribution from the second precursor becomes more apparent rather than being obscured by the greater intensity of the contribution of the first precursor. Determining contributions of one or more additional precursors can be advantageously used to filter out interfering signals and more accurately infer a monoisotopic precursor m / z.

[0113] FIG. 11 is a flowchart of a method 600 of identifying a precursor m / z. In embodiments, the method 600 is performed by a data processing system, such as data processing system 200 of FIG. 3.

[0114] At operation 602, an encoding of mass spectrum data is obtained. The encoding may be of a scanning ion transmission window dimension of mass spectrum data. In embodiments, the encoding includes, for a unique product ion, a position of awindow in which the product ion appears and a sum of counts or intensities for a group of windows spanning the window in which the unique product ion appears.

[0115] At operation 604, a rising linear function and a falling linear function are fit to a response profile of the mass spectrum data. In some embodiments, an apex of the response profile bisects the rising and falling linear functions and may be excluded from at least one of the rising linear function and the falling linear function. In some embodiments, the apex is excluded from each of the rising linear function and the falling linear function. Each of the rising linear function and the falling linear function may be calculated using multiple points, e.g., at least two points, at least three points, at least four points, etc. In some embodiments, the multiple points are consecutive points, e.g., two consecutive points, three consecutive points, four consecutive points, etc.

[0116] At operation 606, an intercept between the rising linear function and the falling linear function is determined. At operation 608, the precursor m / z is identified using the intercept. For example, see intercept 358 of FIG.6, which is associated with an example calculated precursor m / z of 342.17.

[0117] In some embodiments, method 600 further includes evaluating the response profile for a confidence measure of the reliability of the response profile for generating a precursor inference. The confidence evaluation of the response profile may be performed prior to, in parallel with, or following the precursor inference determination. Evaluating the response profile for a confidence measure may include determining a symmetry of the response profile and / or otherwise calculating a quality measure. For example, a quality measure may be determined by calculating a rising fit for the rising linear function and a falling fit for the falling linear function and selecting a lower of the rising fit and the falling fit as the quality measure.

[0118] Referring now to FIG. 12, the example computing system 150 includes one or more processors 152, a system memory 158, and a system bus 172 that couples the system memory 158 to the one or more processors 152. The system memory 158 includes RAM (Random Access Memory) 160 and ROM (Read-Only Memory) 162. A basic input / output system that contains the basic routines that help to transfer information between elements within the computing system 150, such as during startup, is stored in the ROM 162. The computing system 150 further includes a mass storage device 164. The mass storage device 164 is able to store software instructions and data. Mass storage device 164 may correspond to storage 204 of FIG. 3. The one or more processors 152 can be one or more central processing units or other processors.

[0119] The mass storage device 164 is connected to the one or more processors 152 through a mass storage controller (not shown) connected to the system bus 172. The mass storage device 164 and its associated computer-readable data storage media provide nonvolatile, non-transitory storage for the computing system 150. Although the description of computer-readable data storage media contained herein refers to a mass storage device, such as a hard disk or solid state disk, it should be appreciated by those skilled in the art that computer-readable data storage media can be any available non- transitory, physical device or article of manufacture from which the central display station can read data and / or instructions.

[0120] Computer-readable data storage media include volatile and non-volatile, removable and non-removable media implemented in any method or technology for storage of information such as computer-readable software instructions, data structures, program modules or other data. Example types of computer-readable data storage media include, but are not limited to, RAM, ROM, EPROM, EEPROM, flash memory or other solid state memory technology, CD-ROMs, DVD (Digital Versatile Discs), other optical storage media, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to store the desired information and which can be accessed by the computing system 150.

[0121] According to various embodiments of the invention, the computing system 150 may operate in a networked environment using logical connections to remote network devices through the network 148. The network 148 is a computer network, such as an enterprise intranet and / or the Internet. The network 148 can include a LAN, a Wide Area Network (WAN), the Internet, wireless transmission mediums, wired transmission mediums, other networks, and combinations thereof. The computing system 150 may connect to the network 148 through a network interface unit 154 connected to the system bus 172. It should be appreciated that the network interface unit 154 may also be utilized to connect to other types of networks and remote computing systems. The computing system 150 also includes an input / output controller 156 for receiving and processing input from a number of other devices, including a touch user interface display screen, or another type of input device. Similarly, the input / output controller 156 may provide output to a touch user interface display screen or other type of output device.

[0122] As mentioned briefly above, the mass storage device 164 and the RAM 160 of the computing system 150 can store software instructions and data. The software instructions include an operating system 168 suitable for controlling the operation of thecomputing system 150. The mass storage device 164 and / or the RAM 160 also store software instructions, that when executed by the one or more processors 152, cause one or more of the systems, devices, or components described herein to provide functionality described herein. For example, the mass storage device 164 and / or the RAM 160 can store software instructions that, when executed by the one or more processors 152, cause the computing system 150 to receive and execute managing network access control and build system processes.

[0123] The software instructions further include one or more software applications 166. Software applications may include dedicated systems and algorithms for performing specific tasks or actions or providing specific interfaces. One or more of data processing system 200 and / or one or more component of data processing system 200 may be encompassed by software applications 166.

[0124] Illustrative examples of the systems and methods described herein are provided below. An embodiment of the system or method described herein may include any one or more, and any combination of, the aspects described below.

[0125] Aspect 1. A method of determining a confidence measure of precursor inference in mass spectrum data, the method including: obtaining mass spectrum data of a set of overlapping transmission windows; determining a response profile of the mass spectrum data across the set of overlapping transmission windows; evaluating the response profile using response profile metrics; and reporting the confidence measure of the precursor inference using the response profile and the evaluating.

[0126] Aspect 2. The method of aspect 1, wherein the mass spectrum data includes one of a raw mass spectrum acquisition and preprocessed mass spectrum data.

[0127] Aspect 3. The method of aspect 2, wherein preprocessing of the preprocessed mass spectrum data includes one or more of encoding and binning.

[0128] Aspect 4. The method of any one of aspects 1-3, wherein the precursor inference is a predetermined precursor candidate.

[0129] Aspect 5. The method of aspect 4, wherein the response profile metrics are based on one or a combination of an expected response profile and an experimental intensity trace.

[0130] Aspect 6. The method of aspect 5, wherein the expected response profile is based on one or a combination of an expected precursor m / z, an expected precursor isotope pattern, and an expected acquisition engine response.

[0131] Aspect 7. The method of aspect 6, wherein the mass spectrum data includes an MSI scan and the expected precursor m / z is based on one or more of the MSI scan and reference data, wherein the reference data includes one or more or a combination of previously acquired mass spectrum data, experimental library data, and theoretical library data.

[0132] Aspect 8. The method of aspect 5, wherein the expected response profile is a profile across time of flight pulses from one of a raw mass spectrum acquisition, preprocessed mass spectrum data, and an encoding of mass spectrum data.

[0133] Aspect 9. The method of any one of aspects 4-8, wherein the mass spectrum data includes an MSI scan and the predetermined precursor candidate is determined using at least one of the MSI scan and reference data.

[0134] Aspect 10. The method of aspect 9, wherein the reference data includes an expected sample identity based on one or more of previously acquired experimental data and a library of hypothetical compound identities

[0135] Aspect 11. The method of any one of aspects 1-10, wherein the precursor inference is an unknown precursor identified based on the response profile.

[0136] Aspect 12. The method of any one of aspects 1-11, further including: generating a response profile for a plurality of detected product ions in the mass spectrum data; generating a precursor inference for each product ion of the plurality of detected product ions; and grouping two or more product ions of the plurality of detected product ions using a common precursor inference.

[0137] Aspect 13. The method of aspect 12, further including: calculating a representative intensity profile using the grouping of two or more product ions; and scoring each product ion in the grouping of two or more product ions with respect to the representative intensity profile.

[0138] Aspect 14. The method of aspect 12, wherein the response profile for each detected product ion includes calculating a matrix values for each product ion and grouping the two or more product ions of the plurality of detected product ions comprises grouping the two or more product ions having a matrix value within a predetermined tolerance.

[0139] Aspect 15. The method of any one of aspects 12-14, further comprising generating a compound fingerprint using the grouping.

[0140] Aspect 16. The method of any one of aspects 1-15, wherein the response profile metrics include a symmetry of the response profile.

[0141] Aspect 17. The method of any one of aspects 1-16, further including filtering one or more product ions associated with the response profile using the quality measure.

[0142] Aspect 18. The method of aspect 17, wherein filtering the one or more product ions associated with the response profile is performed using an inference precursor ion m / z to remove product ions attributed to a different precursor ion m / z.

[0143] Aspect 19. The method of any one of aspects 1-18, further including filtering a value at each position across the set of overlapping transmission windows.

[0144] Aspect 20. The method of aspect 1, wherein the mass spectrum data is an encoding.

[0145] Aspect 21. The method of aspect 20, further including identifying a slope of a rising linear function and a slope of a falling linear function of the response profile.

[0146] Aspect 22. The method of aspect 20, wherein the response profile metrics include a symmetry of the response profile and the symmetry includes a ratio of the slope of the rising linear function and the slope of the falling linear function.

[0147] Aspect 23. The method of aspect 22, wherein the ratio includes an absolute value of each of the slope of the rising linear function and the slope of the falling linear function.

[0148] Aspect 24. The method of aspect 22 or 23, wherein the response profile metrics include an expected symmetry of 1 for the symmetry.

[0149] Aspect 25. The method of aspect 22 or 23, wherein the response profile metrics further include tolerances for the symmetry determined using experimental data.

[0150] Aspect 26. The method of any one of aspects 22-25, further including filtering one or more product ions associated with the response profile using the symmetry.

[0151] Aspect 27. The method of any one of aspects 20-26, wherein the response profile metrics include a quality measure determined by: calculating a rising fit for a rising linear function and a falling fit for a falling linear function of the response profile; and selecting one of the rising fit and the falling fit as the quality measure.

[0152] Aspect 28. The method of aspect 27, wherein selecting one of the rising fit and the falling fit as the quality measure includes selecting the lower of the rising fit and the falling fit.

[0153] Aspect 29. The method of aspect 27 or 28, wherein each of the rising fit and the falling fit are calculated as a coefficient of determination.

[0154] Aspect 30. The method of any one of aspects 27-30, wherein each of fitting the rising linear function and the falling linear function includes selecting a neighborhood of rising points and a neighborhood of falling point.

[0155] Aspect 31. The method of aspect 30, further including identifying an apex of the response profile which bisects the neighborhood of rising points and the neighborhood of falling points.

[0156] Aspect 32. The method of aspect 31, wherein each of the neighborhood of rising points and the neighborhood of falling points includes at least two points.

[0157] Aspect 33. The method of aspect 31 or 32, wherein each of the neighborhood of rising points and the neighborhood of falling points includes at least three points.

[0158] Aspect 34. The method of any one of aspects 31-33, wherein selecting each of the neighborhood of rising points and the neighborhood of falling points comprises excluding the apex.

[0159] Aspect 35. The method of aspect 34, wherein excluding the apex further includes excluding at least one additional point adjacent to the apex.

[0160] Aspect 36. The method of aspect 34 or 35, wherein excluding the apex further includes excluding at least one additional point adjacent to the apex to either side of the apex.

[0161] Aspect 37. The method of any one of aspects 1-36, wherein the response profile metrics further include a symmetry of the response profile.

[0162] Aspect 38. The method of any one of aspects 20-37, further including calculating a precursor ion m / z using an intercept of the rising linear function and the falling linear function.

[0163] Aspect 39. The method of any one of aspects 1-38, wherein the mass spectrum data includes an encoding and the encoding includes, for at least one product ion, a position of a window in which the at least one unique product ion appears and a sum of counts or intensities for a group of windows spanning the window in which the at least one product ion appears.

[0164] Aspect 40. The method of any one of aspects 1-39, wherein the mass spectrum data includes an encoding and the method further includes calculating a precursor ion m / z for the precursor inference using an intercept of a rising linear function and a falling linear function of the response profile.

[0165] Aspect 41. The method of any one of aspects 1-40, wherein the mass spectrum data includes an encoding and the encoding includes, for at least one production, a position of a window in which the at least one product ion appears and a sum of counts or intensities for a group of windows spanning the window in which the at least one product ion appears.

[0166] Aspect 42. The method of any one of aspects 1-41, further including filtering a sum at each position across the scanning ion transmission window dimension.

[0167] Aspect 43. A method of detecting conflicting relationships between precursors that are captured within the group of windows, the method including: obtaining an encoding of a scanning ion transmission window dimension of mass spectrum data, wherein the encoding includes, for a product ion, a position of a window in which the product ion appears and a sum of counts or intensities for a group of windows spanning the window in which the product ion appears; and filtering the sum at at least one position across the scanning ion transmission window dimension.

[0168] Aspect 44. The method of aspect 43, wherein filtering includes running a subtraction filter.

[0169] Aspect 45. The method of aspect 44, wherein the subtraction filter includes subtracting an average intensity of neighboring positions defined by a neighborhood.

[0170] Aspect 46. The method of aspect 45, wherein the neighborhood includes a number of positions away from the at least one position included in the average.

[0171] Aspect 47. The method of any one of aspects 43-46, further including filtering the sum at each position across the scanning ion transmission window dimension.

[0172] Aspect 48. The method of any one of aspects 43-47, wherein filtering the sum at at least one position across the scanning ion transmission window dimension yields a filtered response profile; and the method further includes filtering one or more product ions associated with the filtered response profile using a precursor ion m / z to remove product ions attributed to a different precursor ion m / z.

[0173] Aspect 49. The method of aspect 48, wherein the precursor ion m / z is determined using the filtered response profile.

[0174] Aspect 50. A method of detecting conflicting relationships between precursors that are captured within the group of windows, the method including: obtaining an encoding of a scanning ion transmission window dimension of mass spectrum data, wherein the encoding includes, for at least one product ion, a position of a window in which the at least one product ion appears and a sum of counts or intensities for a group of windows spanning the window in which the at least one product ionappears; and generating, based on the encoding, a trace of the scanning ion transmission window dimension wherein each sum is substituted with a binary metric for presence of the at least one product ion in each window.

[0175] Aspect 51. The method of aspect 50, further including determining a precursor ion m / z corresponding to the at least one unique product ion as an m / z at which the at least one unique product ion is present in a highest total number of windows.

[0176] Aspect 52. The method of aspect 50 or 51, wherein the encoding further includes a position and a sum for each of at least two unique product ions.

[0177] Aspect 53. The method of aspect 52, wherein the trace includes each of the at least two unique product ions.

[0178] Aspect 54. A method of identifying a precursor m / z, the method including: obtaining an encoding of mass spectrum data; fitting a rising linear function and a falling linear function to a response profile of the mass spectrum data; determining an intercept between the rising linear function and the falling linear function; and identifying the precursor m / z using the intercept.

[0179] Aspect 55. The method of aspect 54, wherein the apex is excluded from at least one of the rising linear function and the falling linear function.

[0180] Aspect 56. The method of aspect 55, wherein the apex is excluded from each of the rising linear function and the falling linear function.

[0181] Aspect 57. The method of any one of aspects 54-56, wherein each of the rising linear function and the falling linear function is calculated using at least two points.

[0182] Aspect 58. The method of aspect 57, wherein each of the rising linear function and the falling linear function is calculated using at least three points.

[0183] Aspect 59. The method of aspect 58, wherein each of the at least three points is three consecutive points.

[0184] Aspect 60. The method of any one of aspects 54-59, further including determining a symmetry of the response profile and generating a confidence measure of the response profile for precursor inference using the symmetry.

[0185] Aspect 61. The method of aspect 60, wherein the symmetry includes a ratio of absolutes values of a slope of the rising linear function and a slope of the falling linear function.

[0186] Aspect 62. The method of any one of aspects 54-61, further including determining a quality measure by: calculating a rising fit for the rising linear functionand a falling fit for the falling linear function; and selecting a lower of the rising fit and the falling fit.

[0187] Aspect 63. The method of any one of aspects 54-62, wherein the encoding is of the scanning ion transmission window dimension includes, for at least one product ion, a position of a window in which the at least one unique product ion appears and a sum of counts or intensities for a group of windows spanning the window in which the at least one product ion appears.

[0188] Aspect 64. The method of any one of aspects 1-63, further comprising determining the precursor inference with reference to an orthogonal data dimension.

[0189] Aspect 65. The method of aspect 64, wherein the orthogonal data dimension is one or more of a time dimension or a mobility dimension.

[0190] Aspect 66. A method of determining a relationship inference of a fragment ion, wherein the relationship inference is indicative of one of a relationship with a particular precursor ion, a relationship with two or more co-isolated precursor ions, a direct relationship with a single precursor ion, and a precursor isotope pattern, the method including: obtaining mass spectrum data of a set of overlapping transmission windows; determining a response profile of the mass spectrum data across the set of overlapping transmission windows; evaluating the response profile using response profile metrics; and reporting a confidence measure of the relationship inference using the response profile and the evaluating.

[0191] Having described the preferred aspects and implementations of the present disclosure, modifications and equivalents of the disclosed concepts may readily occur to one skilled in the art. However, it is intended that such modifications and equivalents be included within the scope of the claims which are appended hereto.

Claims

What is claimed is:

1. A method of determining a confidence measure of precursor inference in mass spectrum data, the method comprising: obtaining mass spectrum data of a set of overlapping transmission windows; determining a response profile of the mass spectrum data across the set of overlapping transmission windows; evaluating the response profile using response profile metrics; and reporting the confidence measure of the precursor inference using the response profile and the evaluating.

2. The method of claim 1, wherein the mass spectrum data comprises one of a raw mass spectrum acquisition and preprocessed mass spectrum data.

3. The method of claim 2, wherein preprocessing of the preprocessed mass spectrum data includes one or more of encoding and binning.

4. The method of any one of claims 1-3, wherein the precursor inference is a predetermined precursor candidate.

5. The method of claim 4, wherein the response profile metrics are based on an expected response profile.

6. The method of claim 5, wherein the expected response profile is based on one or a combination of an expected precursor m / z, an expected precursor isotope pattern, and an expected acquisition engine response.

7. The method of claim 6, wherein the mass spectrum data includes an MSI scan and the expected precursor m / z is based on one or more of the MSI scan and reference data, wherein the reference data includes one or more or a combination of previously acquired mass spectrum data, experimental library data, and theoretical library data.

8. The method of claim 5, wherein the expected response profile is a profile across time of flight pulses from one of a raw mass spectrum acquisition, preprocessed mass spectrum data, and an encoding of mass spectrum data.

9. The method of any one of claims 4-8, wherein the mass spectrum data includes an MSI scan and the predetermined precursor candidate is determined using at least one of the MSI scan and reference data.

10. The method of claim 9, wherein the reference data includes an expected sample identity based on one or more of previously acquired experimental data and a library of hypothetical compound identities11. The method of any one of claims 1-10, wherein the precursor inference is an unknown precursor identified based on the response profile.

12. The method of any one of claims 1-11, further comprising: generating a response profile for a plurality of detected product ions in the mass spectrum data; generating a precursor inference for each product ion of the plurality of detected product ions; and grouping two or more product ions of the plurality of detected product ions using a common precursor inference.

13. The method of claim 12, further comprising: calculating a representative intensity profile using the grouping of two or more product ions; and scoring each product ion in the grouping of two or more product ions with respect to the representative intensity profile.

14. The method of claim 12, wherein the response profile for each detected product ion comprises calculating a matrix values for each product ion and grouping the two or more product ions of the plurality of detected product ions comprises grouping the two or more product ions having a matrix value within a predetermined tolerance.

15. The method of any one of claims 1-14, wherein the response profile metrics include a symmetry of the response profile.

16. The method of claim 1, further comprising filtering one or more product ions associated with the response profile using the quality measure.

17. The method of claim 16, wherein filtering the one or more product ions associated with the response profile is performed using an inference precursor ion m / z to remove product ions attributed to a different precursor ion m / z.

18. The method of any one of claims 1-17, further comprising filtering a value at each position across the set of overlapping transmission windows.

19. A method of detecting conflicting relationships between precursors that are captured within the group of windows, the method comprising: obtaining an encoding of a scanning ion transmission window dimension of mass spectrum data, wherein the encoding includes, for a unique product ion, a position of a window in which the unique product ion appears and a sum of counts or intensities for a group of windows spanning the window in which the unique product ion appears; filtering the sum at at least one position across the scanning ion transmission window dimension to yield a filtered response profile; and filtering one or more product ions associated with the filtered response profile using a precursor ion m / z to remove product ions attributed to a different precursor ion m / z.

20. A method of detecting conflicting relationships between precursors that are captured within the group of windows, the method comprising: obtaining an encoding of a scanning ion transmission window dimension of mass spectrum data, wherein the encoding includes, for at least one unique product ion, a position of a window in which the at least one unique product ion appears and a sum of counts or intensities for a group of windows spanning the window in which the at least one unique product ion appears; andgenerating, based on the encoding, a trace of the scanning ion transmission window dimension wherein each sum is substituted with a binary metric for presence of the at least one unique product ion in each window.

Images