Compounds combinations

Abstract

A combination of compounds is provided. The mass spectrum data includes a plurality of MS1 mass spectra and a plurality of MSN mass spectra, each mass spectrum having a respective associated retention time. The method includes detecting a set of features in the plurality of MS1 mass spectra, where each feature of the set has a respective mass, and where the feature of the set has a corresponding retention time. The method comprises, for each feature of the one or more features of the set: (i) submitting a corresponding MSN mass spectrum to a mass spectrum search engine in order to obtain an identification result for the feature, and (ii) determining a candidate ion type for the feature based on a mass difference between a mass associated with the feature and an expected mass from the identification result. The method includes identifying one or more compounds based on the set of features and the candidate ion type.

Description

Technical Field

[0001] The present invention relates to the field of mass spectrometry and, more particularly, to methods of processing mass spectrometry data. Background Art

[0002] It has been some time since the discovery of soft ionization techniques enabled mass spectrometry analysis of intact small molecules, peptides and proteins. Using these techniques, very limited in-source fragmentation significantly simplifies the spectra obtained. However, individual compounds still appear as a series of mass-to-charge ratio (m / z) signals. In addition to isotopes, these signals include solvent adducts, homo- or hetero-dimers, different charge states and any remaining in-source fragments. The correct "combination" of these multiple m / z signals is an important step in both qualitative and quantitative analysis.

[0003] It is believed that there is still room for improvement in the methods used to process mass spectrometry data. Summary of the invention

[0004] A method for processing mass spectrometry data of a sample is provided. The mass spectrometry data includes multiple MS 1 Mass spectrometry and multiple MS N Mass spectrometry, the multiple MS 1 Mass spectrometry and the multiple MS N Each of the mass spectra has a corresponding associated retention time. The method includes detecting the plurality of MS 1 A set of features in a mass spectrum, wherein each feature of the set has a corresponding mass, and wherein the features of the set have corresponding retention times. The method includes identifying one or more compounds in the sample based on the set of features.

[0005] As described in more detail below, various embodiments provide improved methods of processing mass spectrometry data.

[0006] In an embodiment, the mass spectral data is generated by an analytical instrument including a chromatographic separation device (such as a liquid chromatography (LC) separation device or a gas chromatography (GC) separation device) and a mass spectrometer. The chromatographic separation device can separate the sample in a chromatographic separation scan, and the mass spectrometer can obtain mass spectral data during the chromatographic separation scan. Thus, the multiple MS 1 Mass spectrometry and the multiple MS N Each of the mass spectra will have a corresponding associated retention time, ie, each mass spectrum will be acquired at a corresponding chromatographic retention time during that chromatographic separation scan.

[0007] As described further below, in some embodiments, each MS N The mass spectrum is MS2. However, each MS NAlternatively it may be a higher order fragmentation spectrum, such as an MS3 spectrum. Typically, N is an integer ≥2.

[0008] The mass spectrometry data may be composed of (i.e., may be formed from) at least one sample file, wherein each sample file is data output from a corresponding chromatographic separation scan from a mass spectrometer. The mass spectrometry data of each sample file may include a plurality of MS 1 Multiple MS in mass spectrometry 1 Mass spectrometry and multiple MS N Multiple MS in mass spectrometry N Mass spectra. The mass spectrometry data may include a single such sample file or multiple sample files. In the case of multiple sample files, each sample file may be obtained from a different fractionated chromatographic separation scan of the same sample.

[0009] The method includes detecting the plurality of MS 1 A set of features in a mass spectrum. Each feature in the set may have a corresponding mass and a corresponding retention time associated therewith. The features in the set may have corresponding (e.g., equal within a first tolerance) retention times, but different masses. The method may include detecting the plurality of MS 1 One or more additional feature groups in the mass spectrum, wherein each respective different group has a respective different retention time. Although the processing of only a single feature group is described in detail below, it will be appreciated that each additional feature group may be processed in a similar manner.

[0010] In some embodiments, by first detecting the plurality of MS 1 The features in the mass spectrum are then divided into detected features with corresponding (i.e., equal within a first tolerance) retention times to detect each feature group. 1 The per-file feature is detected in the mass spectrum and then divided across sample files with corresponding masses and retention times to form features that can be used in multiple MS 1 Detection features in mass spectrometry.

[0011] Therefore, the detection of the multiple MS 1 The step of forming a feature group in a mass spectrum may include: for each sample file: detecting a plurality of per-file features in the sample file, each per-file feature having a corresponding mass and a corresponding retention time; forming a plurality of features from the per-file features, each feature having a corresponding mass and a corresponding retention time; and forming a feature group by dividing the features having corresponding retention times.

[0012] In these embodiments, the multiple MS 1Each unique mass-to-charge ratio (m / z) in the mass spectrum is used to construct a chromatogram and a characteristic retention time is determined for each chromatogram to detect each per-file feature in the sample file. The characteristic retention time for the chromatogram can be the retention time at the center or vertex of the chromatogram and can be determined, for example, using a peak detection algorithm or a similar algorithm. The chromatograms can then be divided into sets according to the characteristic retention times of the chromatograms, and a deisotoping algorithm can be applied to each chromatogram set to form a per-file feature group.

[0013] Therefore, each per-file feature is in the deisotopic MS 1 A feature occurring in the mass spectral data, the feature having a unique combination of mass and retention time. Then, each per-file feature of each per-file feature group will have a corresponding mass and a corresponding retention time, and each per-file feature of each group will have a corresponding (i.e., the same within a second tolerance) retention time.

[0014] Likewise, the step of detecting a plurality of per-file features in the sample file may include: 1 The mass spectrometer constructs a plurality of chromatograms, each chromatogram having a corresponding mass-to-charge ratio (m / z); determines a characteristic retention time for each chromatogram; divides the chromatograms having the corresponding characteristic retention time into one or more sets of chromatograms; and applies a deisotoping algorithm to each set of chromatograms to form a per-file feature group.

[0015] As described above, the features in each feature group may have equal retention times within a first tolerance, and the per-file features in each per-file feature group may have equal retention times within a second tolerance. The second tolerance may be smaller than the first tolerance. As further described below, this has the effect of properly accounting for retention time differences between different sample files.

[0016] Therefore, the step of dividing features having corresponding retention times may include dividing features having retention times that are equal within a first tolerance; the step of dividing chromatograms having corresponding characteristic retention times may include dividing chromatograms having retention times that are equal within a second tolerance; and the second tolerance may be smaller than the first tolerance.

[0017] In some embodiments, the method includes: for each feature in one or more features in the feature set: (i) submitting the corresponding MS to a mass spectrometry search engine; NThe method further comprises the steps of: (i) determining a mass spectrum of the feature to obtain an identification result for the feature, and (ii) determining a candidate ion type (such as a candidate adduct ion type) for the feature based on any mass difference between the mass associated with the feature and the expected mass from the identification result. In this case, the step of identifying one or more compounds in the sample can be based on both the set of features and the candidate (adduct) ion types.

[0018] In these embodiments, steps (i) and (ii) may be performed on the corresponding MS in the group. N The data is performed based on those features available in the mass spectrometry data. In the case of multiple sample files, the corresponding MS N Mass spectrometry. Therefore, the MS corresponding to the characteristic N The mass spectrum may be the MS of any of the per-file features corresponding to the feature N Spectra (from multiple MS N mass spectrometry).

[0019] As described above, the mass spectrometry data may include only a single sample file, in which case each of the plurality of features is formed from one per-file feature of the plurality of per-file features.

[0020] In these embodiments, the step of identifying one or more compounds based on the set of features may include:

[0021] (i) determining one or more feature clusters, wherein each feature cluster includes one or more of the group

[0022] features and may correspond to the corresponding (single) compounds;

[0023] (ii) determining one or more arrangement structures of the feature cluster for the feature group, wherein each arrangement

[0024] The structure includes one or more non-conflicting feature clusters; and

[0025] (iii) selecting a preferred arrangement structure for the feature group from one or more arrangement structures of the feature cluster;

[0026] and then:

[0027] One or more compounds are identified based on the preferred arrangement of the characteristic clusters.

[0028] However, in certain embodiments, the mass spectrometry data includes a plurality of sample files, and each of the plurality of features is formed by partitioning the per-file features across sample files having a corresponding mass and a corresponding retention time. In this case, for each feature group, there is a corresponding per-file feature group in each sample file of the plurality of sample files.

[0029] The step of identifying one or more compounds based on the set of features may then include:

[0030] For each per-file feature group in the corresponding sample file:

[0031] (i) determining one or more per-file feature clusters, wherein each per-file feature cluster includes the group

[0032] One or more per-file features and possibly corresponding to corresponding compounds;

[0033] (ii) determining one or more arrangement structures of each file feature cluster for each file feature group, wherein

[0034] Each arrangement structure includes one or more non-conflicting per-file feature clusters; and

[0035] (iii) selecting from one or more per-file feature clusters for the per-file feature group

[0036] Optimize the arrangement structure;

[0037] And then, based on the preferred arrangement structure for the plurality of sample files:

[0038] (iv) determining one or more arrangement structures of feature clusters for the feature group, wherein each feature cluster

[0039] comprising one or more features of the feature set and possibly corresponding to a corresponding compound,

[0040] and wherein each arrangement structure includes one or more non-conflicting feature clusters; and

[0041] (v) selecting a preferred arrangement for the feature group from one or more arrangements of the feature cluster;

[0042] and then

[0043] One or more compounds are identified based on the preferred arrangement of the characteristic clusters.

[0044] As described in more detail below, this two-step process of first determining a preferred arrangement for each sample file and then resolving any conflicts between preferred arrangements from different sample files has the effect of increasing the reliability and consistency of compound identification when multiple sample files are present.

[0045] In some embodiments, for each per-file feature group in a corresponding sample file, the step of determining one or more per-file feature clusters may include: assigning one or more candidate ion types to each per-file feature of the group; determining one or more candidate relationships between the per-file features of the group; and resolving any conflicts between the candidate ion types and the candidate relationships for the group.

[0046] In these embodiments, the step of assigning one or more candidate ion types to each per-file feature of the group may include assigning one or more candidate ion types from a plurality of different categories to each per-file feature of the group. The candidate ion type categories may include, for example, (i) identified ion types; (ii) user-defined base ion types; (iii) default ion types; and (iv) in-source fragment ion types.

[0047] Likewise, the step of determining one or more candidate relationships (or "transitions") between the set of per-file features may include determining one or more candidate relationships from a plurality of different categories of relationships. The categories of relationships may include, for example, (i) in-source fragment relationships; and (ii) adduct relationships.

[0048] Thus, for example, each per-file feature of the group can have a corresponding charge, and the step of assigning one or more candidate ion types to each per-file feature of the group can include: assigning an identified ion type to any per-file feature of the group corresponding to the feature for which an identification result was obtained; and / or assigning a user-defined alkali ion type or a default ion type to each per-file feature of the group based on the corresponding charge of the per-file feature.

[0049] Additionally or alternatively, the step of assigning one or more candidate ion types to each per-file feature of the group may include assigning an in-source fragment ion type to any per-file feature of the group, the mass of which corresponds to the mass of an expected in-source fragment of another per-file feature in the group.

[0050] In this case, the step of determining one or more candidate relationships between the per-file features of the group may include determining the intra-source fragmentation relationship between the per-file feature of the group and another per-file feature of the group when the quality of the per-file feature corresponds to the quality of the expected intra-source fragmentation of another per-file feature (or, in other words, when the quality of the other per-file feature corresponds to the quality of the expected intra-source fragmentation of the per-file feature).

[0051] In these embodiments, the MS corresponding to each file feature can be N Mass spectrometry (from multiple MS NA mass spectrum) is used to determine the masses of one or more expected in-source fragments that characterize each file.

[0052] Alternatively, the method may include providing the quality of one or more expected in-source fragments of a feature as part of the recognition result for the feature. In this case, the quality of one or more expected in-source fragments of each file feature may be derived from the provided quality.

[0053] In addition, one or more MS configured to simulate in-source fragmentation can be used. N The mass spectrometer determines the mass provided as part of the identification result. As described in more detail below, by using a custom MS in this manner N Mass spectrometry can significantly improve the identification of in-source fragmentation.

[0054] In some embodiments, determining one or more candidate relationships between the set of per-file features may include determining one or more candidate adduct relationships between the set of per-file features based on allowed mass deviations between masses of the set of per-file features.

[0055] In an embodiment, once all possible candidate ion types and candidate relationships have been determined for each file feature set, any conflicts are resolved.

[0056] In some embodiments, the step of resolving any conflicts between the candidate ion type and the candidate relationship may include: removing any candidate relationship that conflicts with a per-file feature assigned to the identified ion type (e.g., removing any candidate adduct relationship and / or removing any candidate intra-source fragment relationship that conflicts with a per-file feature assigned to the identified ion type); and / or removing any illegal candidate intra-source fragment relationships.

[0057] As described above, in some embodiments, the method includes determining one or more per-file feature clusters for each per-file feature group in the corresponding sample file. Each such determined cluster can potentially correspond to a corresponding compound.

[0058] Thus, in an embodiment, the step of determining one or more per-file feature clusters may include: determining most or all possible per-file feature clusters; and then removing any invalid clusters; and / or removing any clusters that do not include a per-file feature assigned to a user-defined alkali ion.

[0059] As described above, the method may include determining one or more arrangements of clusters (per file feature or features) for each group, and then selecting one of the arrangements as a preferred arrangement. Each such arrangement includes one or more clusters that do not conflict with each other. Any cluster that does not have a conflicting cluster can be used immediately in a preferred arrangement without further processing.

[0060] Thus, the step of determining one or more arrangements of clusters (per file feature or features) may comprise determining whether a cluster conflicts with one or more other clusters; and when it is determined that a cluster does not conflict with any other cluster, using the cluster in a preferred arrangement of clusters for the group.

[0061] Conversely, where there are conflicting clusters, these conflicting clusters must be resolved against each other. In an embodiment, this is accomplished by determining a plurality of different arrangements of clusters, assigning a score to each arrangement, and selecting the arrangement with the highest score as the preferred arrangement.

[0062] Thus, the step of selecting a preferred arrangement of clusters (per file feature or features) from one or more arrangements of clusters may comprise: determining a score for each arrangement of clusters; and selecting the arrangement with the highest score.

[0063] The step of determining a score for each arrangement structure of a cluster may include: determining a cluster score for each cluster in the arrangement structure by: (i) assigning a weight factor to each candidate ion type assignment for the cluster, (ii) assigning a relationship score to each candidate relationship for the cluster, and (iii) calculating a cluster score for the cluster by dividing the sum of the weight factor and the relationship score by the number of features or per-file features in the cluster; and determining the score for each arrangement structure by dividing the sum of the cluster scores of the arrangement structure by the number of clusters in the arrangement structure.

[0064] Another aspect provides a mass spectrometry analysis method, the method comprising: analyzing a sample to obtain a plurality of MS 1 Mass spectrometry and multiple MS N mass spectral data of mass spectra, each mass spectrum having a corresponding associated retention time; and processing the mass spectral data using the above method.

[0065] Another aspect provides a non-transitory computer-readable storage medium storing computer software code which, when executed on a processor, performs the method described above.

[0066] Another aspect provides a control system for an analytical instrument such as a mass spectrometer, the control system being configured to cause the analytical instrument to perform the method described above.

[0067] Another aspect provides an analytical instrument, such as a mass spectrometer, comprising a control system as described above. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Various embodiments will now be described in more detail with reference to the accompanying drawings, in which:

[0069] Figure 1 schematically illustrates a mass spectrometer that may operate according to an embodiment;

[0070] Figure 2 Schematically illustrates a method according to an embodiment;

[0071] Figure 3 A representation of a main relationship graph including a plurality of features is shown;

[0072] Figure 4 showing a representation of a per-file relationship graph including a plurality of per-file features;

[0073] Figure 5A shows a representation of per-file features with identified ion types, Figure 5B shows a representation of per-file features that have been labeled as either of two different user-defined alkali ion types, Figure 5C shows a representation of the per-file features that have been tagged with the default ion type, Figure 5D shows a representation of per-file features that have been labeled as either one of two different user-defined alkali ion types or as in-source fragment ions;

[0074] Figure 6 shows a representation of a per-file relationship graph including two per-file features and including various possible ion types and transitions;

[0075] Fig. 7A and Figure 7B shows the process of removing invalid transitions from the initial per-file relationship graph;

[0076] Figure 8 shows a representation of a per-file relationship graph including eight per-file features and including various possible ion types and transitions;

[0077] Fig. 9 Shows from Figure 8 Representation of various possible clusters formed by each file feature;

[0078] Fig.10 shows the process of removing invalid clusters from the initial set of possible clusters;

[0079] Fig.11shows the removal of invalid clusters from the initial set of possible clusters

[0080] The process of isolating clusters;

[0081] Fig.12 shows the invalid clusters after they have been removed from Figure 8 Representation of various possible clusters formed by each file feature;

[0082] Fig.13 Shows from Fig.12 The process of determining various possible explanations for each file feature group from various possible clusters;

[0083] Fig.14 The final interpretation of the per-file feature set determined according to the implementation scheme is shown;

[0084] Fig.15 shows the process of parsing the fragments; and

[0085] Fig.16A The experimental results using the conventional combination method are shown, and Fig. 16B Experimental results using the combined approach of various embodiments are shown. DETAILED DESCRIPTION

[0086] Figure 1 Schematically illustrates analytical instruments, such as a mass spectrometer, that can be used in conjunction with the methods described herein. Figure 1 As shown, the instrument includes an ion source 10 , a mass filter 20 , a fragmentation device 30 and a mass analyzer 40 .

[0087] The ion source 10 is configured to generate ions from a sample. The ion source 10 may be coupled to a chromatographic separation device (not shown), such as a liquid chromatography (LC) separation device, a gas chromatography (GC) separation device, or a capillary electrophoresis separation device, so that the sample ionized in the ion source 10 comes from the separation device. The ion source 10 may be any suitable ion source, such as an electrospray ionization (ESI) ion source, an atmospheric pressure ionization (API) ion source, a chemical ionization ion source, an electron impact (EI) ion source, or the like.

[0088] The mass filter 20 is arranged downstream of the ion source 10 and is configured to receive ions from the ion source 10. The mass filter 20 is configured to filter the received ions according to the mass-to-charge ratio (m / z) of the received ions. The mass filter 20 can be configured so that the received ions with m / z within the m / z transmission window of the mass filter are transmitted forward by the mass filter, while the received ions with m / z outside the m / z transmission window are attenuated by the mass filter, i.e., not transmitted forward by the mass filter. The width and / or center m / z of the transmission window are controllable (variable) by, for example, appropriately controlling the RF voltage and / or DC voltage applied to the electrodes of the mass filter 20. Therefore, for example, the mass filter 20 can be operated in a transmission operation mode, whereby most or all of the ions within a relatively wide m / z window are transmitted forward by the mass filter 20, and can be operated in a filtering operation mode, whereby only the ions within a relatively narrow m / z window (centered on the desired m / z) are transmitted forward by the mass filter 20. The mass filter 20 may be any suitable type of mass filter, such as a quadrupole mass filter.

[0089] The fragmentation device 30 is arranged downstream of the mass filter 20 and is configured to receive most or all of the ions transmitted by the mass filter 20. The fragmentation device 30 can be configured to selectively fragment some or all of the received ions, i.e., to produce fragment ions. The fragmentation device 30 can be capable of operating in a fragmentation operating mode, whereby most or all of the received ions are fragmented to produce fragment ions (which can then be transmitted forward from the fragmentation device 30), and can be capable of operating in a non-fragmentation operating mode, whereby most or all of the received ions are transmitted forward without being (intentionally) fragmented. It is also possible to achieve a non-fragmentation operating mode by allowing ions to bypass the fragmentation device 30. The fragmentation device 30 may also be capable of operating in one or more intermediate operating modes, for example, whereby the degree of fragmentation is controllable (variable). The fragmentation device 30 may also be capable of operating in a higher order (MS N ) fragmentation operation mode, for example whereby the fragment ions are further fragmented one or more times by the fragmentation device 30.

[0090] The fragmentation device 30 may be any suitable type of fragmentation device, such as, for example, a collision induced dissociation (CID) fragmentation device, an electron induced dissociation (EID) fragmentation device, a photodissociation fragmentation device, etc. Many other types of fragmentation are possible.

[0091] In some embodiments, the fragmentation device 30 is a collision induced dissociation (CID) fragmentation device. Therefore, the fragmentation device may include a collision cell that may be filled with, for example, a collision gas maintained at a relatively high pressure. The ions may be selectively fragmented in the collision cell by controlling (changing) the kinetic energy at which the ions enter the collision cell. In a fragmentation mode of operation, the ions may be accelerated so that the ions enter the collision cell with a relatively high kinetic energy, which may cause most or all of the accelerated ions to fragment. In a non-fragmentation mode of operation, the ions may be caused to enter the collision cell with a relatively low kinetic energy, which may not be sufficient to cause most or all of the ions to fragment. In an intermediate mode, the ions may be caused to enter the collision cell with an intermediate kinetic energy.

[0092] The mass analyzer 40 is arranged downstream of the fragmentation device 30 and is configured to receive ions from the fragmentation device 30. Therefore, depending on the operating mode of the fragmentation device 30, the mass analyzer 40 can receive unfragmented precursor ions and / or fragment ions. The mass analyzer 40 is configured to analyze the received ions in order to determine their mass-to-charge ratio (m / z) and / or mass, that is, to generate a mass spectrum of the ions. The mass analyzer 40 can be any suitable type of mass analyzer, such as an ion trap mass analyzer, an electrostatic orbital trap mass analyzer (such as the Orbitrap manufactured by Thermo Fisher Scientific), or a mass analyzer having a mass spectrum of ions. TM FT mass analyser) or a time-of-flight (ToF) mass analyser such as a multi-reflection time-of-flight (MR-ToF) mass analyser.

[0093] It should be noted that Figure 1 Only schematically, and the instrument may and in embodiments does include any number of additional components in one or more additional components. Typically, the instrument may include one or more ion transfer stages arranged between any of the components shown, such as including an atmospheric pressure interface and / or one or more ion guides, lenses, and / or other ion optical devices configured so that some or all of the ions can be appropriately transmitted through the instrument. The ion transfer stage may include any suitable number and configuration of ion optical devices, such as optionally including one or more ion guides, lenses, and / or other ion optical devices.

[0094] In some embodiments, the instrument may include more than one mass analyser.For example, the instrument may be a dual mass analyser hybrid mass spectrometer of the type described in EP 3,410,463, the contents of which are incorporated herein by reference.

[0095] Likewise Figure 1As shown, the instrument is under the control of a control unit 50 (such as a suitably programmed computer), which controls the operation of the various components of the instrument and, for example, sets the voltages to be applied to the various components of the instrument. The control unit 50 can also receive and process data from the various components comprising the analyzer in various embodiments.

[0096] The instrument may be capable of operating in various operating modes. In particular, the instrument may be a tandem mass spectrometer capable of operating in MS1 ​​operating mode and MS2 operating mode.

[0097] In the MS1 (or "full mass scan") mode of operation, the mass filter 20 operates in its transmission mode of operation and the fragmentation device 30 operates in its non-fragmentation mode of operation, for example so that a wide m / z range (e.g., the full mass range) of unfragmented ("precursor" or "parent") ions is analyzed by the analyzer 40 to produce an MS1 ​​spectrum.

[0098] In the MS2 operating mode, the mass filter 20 operates in its filtering operating mode and the fragmentation device 30 operates in its fragmentation operating mode, for example so that precursor ions of a selected narrow m / z range are fragmented and the resulting fragment ("product" or "daughter") ions are analyzed by the analyzer 40 to produce an MS2 spectrum.

[0099] The instrument may also be capable of operating in one or more higher order fragmentation operating modes, such as, for example, an MS3 operating mode, whereby precursor ions are fragmented, at least some of the resulting fragment ions are themselves fragmented, and second generation fragment ions ("grandchild ions") are analyzed by analyzer 40 to produce an MS3 spectrum. In general, the instrument may be operated in any order of fragmentation operating mode, i.e., in an MS N Operation mode operation, where N ≥ 2.

[0100] The method of operating the analytical instrument involves providing a sample to a chromatographic (e.g., LC or GC) separation device so that the sample is chromatographically separated, ionizing the eluent from the chromatographic separation device in the ion source 10, and analyzing the resulting ions. Different compounds within the sample experience different retention times (RT) within the chromatographic separation device, and therefore elute (and are ionized) from the chromatographic separation device at different times. The chromatographic separation device typically takes tens of seconds or minutes to complete each chromatographic separation scan.

[0101] During each chromatographic separation scan, multiple MS2 spectra (or more generally, multiple MS2 spectra) can be obtained by successively changing the center of the (narrow) m / z window of the mass filter between each of a plurality of different m / z values. N spectrum), for example, to sequentially select (and fragment) each of a plurality of different precursor ions having correspondingly different m / z.

[0102] In a data-dependent acquisition (DDA) mode of operation, a plurality of different m / z values ​​may correspond to a plurality of different precursor ions identified from the corresponding MS1 data (i.e., a full mass scan). Thus, a typical data-dependent acquisition (DDA) method involves repeatedly performing the following steps during a chromatographic separation scan: (i) obtaining an MS1 ​​spectrum within the m / z range of interest; (ii) identifying one or more precursor ions of interest in the MS1 spectrum; and (iii) obtaining an MS2 (or MS 500 mass) spectrum for each identified precursor ion of interest. N Step (iii) comprises, for each of the identified precursor ions: separating the precursor ions using a mass filter 20, fragmenting the separated precursor ions in a fragmentation device 30, and mass analyzing the fragment ions using a mass analyzer 40.

[0103] In data independent acquisition (DIA) MS2 (or MS N ) mode of operation, a plurality of different m / z values ​​can be taken from a predetermined (fixed) list, i.e. without reference to MS1 data. For example, a narrow m / z isolation window can be continuously stepped over the entire m / z range of interest, e.g. as described in EP 3,410,463.

[0104] In any case, each chromatographic separation scan performed by the analytical instrument will generate a sample file. Each sample file includes data on a plurality of MS1 spectra, each MS1 spectrum having an associated retention time; and data on a plurality of MS2 (or MS N ) spectrum data, each MS2 (or MS N ) spectra have associated retention times and associated mass filter isolation windows or precursor ion m / z. The prepared samples can be fractionated, and a chromatographic separation scan can be performed on each fraction to generate multiple such sample files for the same sample.

[0105] The repetition rate of the DDA / DIA method can be fast enough to sample each chromatographically separated compound of interest multiple times during its chromatographic elution. Thus, for each sample file, a chromatographic trace can be constructed for each unique m / z of interest from multiple MS1 spectra. Such a chromatogram typically exhibits peaks corresponding to the chromatographic elution peaks of each chromatographically separated compound.

[0106] The center (e.g., apex) retention time (RT) of each chromatogram can then be determined, for example using a suitable peak detection algorithm or the like. Chromatograms having the same (e.g., within a certain tolerance) center retention time (RT) can be grouped together to produce one or more sets of multiple m / z, where all m / z within a set have been determined to have the same retention time (RT).

[0107] Each such set of multiple m / z can be generated by multiple different compounds co-eluting from the chromatographic separation device, and can be very complex and difficult to interpret. Therefore, it is necessary to "group" each set of multiple m / z into one or more clusters, where each cluster is a subset of m / z from the set (or, where appropriate, the complete set) of multiple m / z determined to have the same RT, where all m / z in the cluster belong to the same single compound. In other words, each single chromatographically separated compound can generate multiple clusters of m / z (with the same RT) in the MS1 data, creating complexity in the MS1 data when multiple co-eluting compounds are present.

[0108] A single compound can produce multiple different m / z in MS1 ​​(at the same RT) due to, for example, different charge states (z) and different isotopes. A collection of isotopes appears in the MS1 data as a series of characteristic peaks, spaced apart from each other by 1 m / z for singly charged species, 1 / 2 m / z for doubly charged species, 1 / 3 for triply charged species, etc. This understanding means that the MS1 data can be "deisotoped", for example by dividing those m / z corresponding to a collection of isotopes (in a collection of multiple m / z determined to have the same RT) together into so-called "per-file features", where each per-file feature has a single characteristic m / z (e.g., corresponding to the lightest isotope). This understanding also means that the correct charge state can be determined for each such per-file feature (based on the m / z separation between the isotopic peaks). (Any "singlet" appearing in the MS1 data (i.e., a peak appearing without any corresponding isotope - typically a low abundance peak where the corresponding isotope is below the detection limit) can be assumed to be singly charged (or have some other default charge depending on sample type, etc.) Knowledge of the charge state (z) and mass-to-charge ratio (m / z) in turn means that a single characteristic mass (m) can be determined for each per-file feature. Algorithms for such deisotoping and charge state determination are known in the art.

[0109] However, despite such deisotoping algorithms, there may still be multiple per-file features for each compound. In other words, once each set of multiple m / z has been deisotoped, multiple per-file feature groups may remain (where each such multiple per-file feature group may be generated by a single compound or by multiple different co-eluting compounds), and therefore each multiple per-file feature group needs to be further combined into one or more clusters, where all per-file features in a cluster belong to the same single compound.

[0110] There can be several different reasons for having multiple per-file features for a single compound. In particular, this can be due to:

[0111] (i) Solvent adducts; that is, adducts during ionization can result in peaks in MS1 ​​with relative

[0112] The neutral mass M of a compound has a characteristic mass difference. For example, in the protonated ion [M+H] +

[0113] With a mass difference of about 1, the ammonia adduct [M+NH4] + With a mass difference of about 18 etc.

[0115] (ii) Homodimers or heterodimers; that is, two proteins linked together. These can again be expressed as

[0116] The quality of the features in MS1 ​​is now poor.

[0117] (iii) In-source fragmentation; that is, unintentional fragmentation of ions can result in fragment ions appearing in the MS1 spectrum

[0118] (not just the expected MS2 spectrum), making it difficult to interpret the MS1 spectrum.

[0119] Likewise, where there are multiple sample files for the same sample, there may be multiple "features" for each compound (where a feature is a collection of multiple corresponding per-file features from multiple sample files). Ideally, all corresponding per-file features for a feature would be identical in terms of m / z and RT, but in practice RT may vary slightly between different chromatographic separation runs. Therefore, it is desirable to combine each set of multiple features into one or more clusters, where all features in a cluster belong to the same single compound.

[0120] Various embodiments are directed to methods for grouping such features or per-file features into clusters, ie, determining which features or per-file features relate to the same compound.

[0121] Known methods for performing such compound combinations are based solely on the expected mass shifts between features per file. However, the inventors have now recognized that these methods have several problems:

[0122] (i)(a) Compounds with intrinsic charges cannot be correctly identified. Some compounds have intrinsic charges.

[0123] charges are present, and these do not usually form additional adduct ions.

[0124] There is no observable mass shift (between at least two different ions) in

[0125] Provide hints for their correct assignment (i.e., it is not possible to determine from MS1 data alone whether the ions are

[0126] For example, protonated ions [M+H] + Or is it an ammonia adduct ion [M+NH4] + etc., which needs to be known to be able to reliably determine the neutral mass M of a compound). In known methods, these "isolated ions" may be incorrectly assigned, for example using default ion definitions (e.g., protonated ions [M+H] for electrospray ionization). + As described further below, various embodiments provide a method for comparing the performance of a database to a standard database (such as, for example, mzCloud TM Database) to search for fragmentation of compounds (e.g., MS2 or MS N ) data to perform preliminary identification of compounds to solve this problem.

[0127] (i)(b) Relatedly, compounds that do not form a (de)protonated ion are often misidentified. Some compounds prefer to produce a protonated ion that may be different from the default ion (e.g., different from the protonated ion [M+H] generated by electrospray ionization). + ) type of ion. For example, some lipids tend to form only sodium adduct ions [M+Na] + This creates the same problem as described above. As described further below, various embodiments again address this challenge by performing a preliminary identification of a compound by searching the compound's fragmentation data against a standard database.

[0128] (ii) Unidentified compound-specific in-source fragmentation. Many compounds undergo a small amount of fragmentation during ionization (or elsewhere in the instrument during an MS1 ​​scan), so their fragments appear in the MS1 spectrum. These fragment ions may then be incorrectly processed by the combined algorithm as additional MS1 features, producing false positive hits. An example of this is PEG (polyethylene glycol) that has lost one or more EG (ethylene glycol) units. When analyzed, it looks as if the sample contains both PEGn6 and PEGn5, although the latter is actually a fragment generated internally in the instrument. As described below, various embodiments obtain a baseline MS2 or MS1 spectrum for each feature. N Data (e.g., from MS2 or MS N Data or by making the identification step (e.g., mzCloud TM ) from low energy MS2 or MS N The collection of spectra returns standard MS2 or MS N spectrum (which can be configured to simulate in-source fragmentation)), and then search the MS1 data for MS2 or MS N The pieces in the data match the pieces to solve the puzzle.

[0129] (iii) Inconsistent adduct type assignments across multiple sample files. There are several reasons why different bins of the same sample may produce different results with respect to adduct type assignments. This may be due, for example, to intensity / abundance variations between bins causing one or more of the adducts to be below the detection limit, resulting in the entire adduct cluster being assigned differently. As described below, embodiments address this issue by placing multiple, potentially competing, individual bin interpretations (or "arrangements") into a final graph that is resolved at the end of the process.

[0130] Figure 2 A method according to an embodiment is shown. Figure 2 As shown, in the first step (step 100), per-file features are detected in separate sample files (in the manner described above), and then these features are divided into features across multiple sample files by matching their RT and m / z values. These merged features are inserted into the main relationship graph 510, for example as nodes.

[0131] Figure 3 An example illustration of such a master relationship diagram 510 is shown. Figure 3 In , each shaded circle (node) represents an identified feature.

[0132] return Figure 2 In step 200, if MS2 (or MS N ) data can be used for any feature detected (MS2 or MS within any sample file from the same sample N Data), the data is submitted to the database search algorithm for identification. Any suitable database search algorithm may be used, such as, for example, mzCloud TM Spectral library. Possible adduct type assignments are determined from the returned results (which will include the precursor ion neutral mass M). In particular, the mass shift between the measured mass and the retrieved identification mass can be used to calculate the ion type of the feature. For example, if the mass measured in MS1 ​​is (M+18), then the feature is identified as an ammonia adduct [M+NH4] + , and this information is inserted into the main relationship graph 510 as a possible identification of the ion type.

[0133] This process not only allows high confidence assignment of ion types to features, but also provides a means to correctly identify compounds with intrinsic charges. Possible ion types can be obtained directly from identification metadata or from one or more predefined lists, or can be automatically generated for a given charge state, e.g. by (de)protonation, if necessary.

[0134] Thus, in an embodiment, library search 200 is performed prior to final compound combination at step 500. The result of search 200 is not considered a definitive identification of a feature (as is conventional), but rather is considered a strong possibility that must be resolved later against any other competing interpretations and identifications (e.g., from other sample files).

[0135] implement Figure 2 Steps 300 and 400 are performed to partition the per-file features for each per-file feature group (at the same RT) into possible clusters of related per-file features. To address issue (iii) above, the process works on a per-file basis, then all possible interpretations are added to the master graph 510, and any conflicts between interpretations from different sample files are resolved in step 500 (this will be described further below).

[0136] In step 300, the per-file features in each sample file are partitioned at step 300 by matching RT values ​​(e.g., in the manner described above) and optionally matching one or more other attributes (e.g., peak shape, etc.). Because the process works on a per-file basis at this stage, very narrow tolerances can be used to partition per-file features by RT. As shown, feature mapping is performed at step 320.

[0137] Figure 4 An example illustration of the resulting per-file relationship graph is shown. In this per-file context, Figure 4 Each shaded circle in represents a per-file feature in a per-file feature group that is determined to have the same RT.

[0138] exist Figure 2 In step 400, the partitioned per-file features are analyzed to determine all possible relationships between the various per-file features in each group.

[0139] This involves first provisionally assigning all possible ion types to each per-file feature of the set. FIG. 5A to FIG. 5D The various possible classes of ion types are shown, including the identification ion ( Figure 5A ), alkali ions ( Figure 5B ), default ion( Figure 5C ) and fragment ions ( Figure 5D ). At this stage, each per-file feature may have multiple possible provisional ion type assignments, which will be resolved later in the process.

[0140] The ion type identification information (from step 200) is added to those per-file features for which corresponding features were identified in step 200. Thus, for example, Figure 5A represents the ion identified as protonated ion [M+H] in step 200 +Per-file features of features.

[0141] Each per-file feature is also temporarily assigned one or more user-defined base ions ( Figure 5B ), depending on the known charge of the per-file feature (where the charge of each per-file feature is known from the deisotope algorithm as described above (or from some other pre-processing of the data)), or is temporarily assigned as a default ion ( Figure 5C ), where the user has not defined an alkali ion for an ion of known charge for that per-file feature. The user-defined alkali ions may be set as desired by the user, e.g., depending on sample chemistry and / or configuration of the ion source, etc. As further described below, in some embodiments, at least one user-defined alkali ion may be required to be present in a cluster in order for the cluster to be considered valid. The default ion may be the most common ion formed during ionization (e.g., by (de)protonation) of an ion having a charge per-file feature. For example, for electrospray ionization, the default ion may be set to the protonated ion [M+H]+ for a singly positively charged ion, the protonated ion [M+2H]+ for a doubly positively charged ion, and the protonated ion [M+2H]+ for a doubly positively charged ion. 2+ etc.

[0142] So, for example, Figure 5B Represents a single positively charged per-file feature that has been temporarily assigned to either of two user-defined alkali ions, i.e., the protonated ion [M+H] + or sodium adduct ion [M+Na]+. Figure 5C Indicates that it has been temporarily assigned the default doubly positively charged ion, i.e. [M+2H] 2+ The double positively charged per-file feature.

[0143] Return to Figure 2 In step 410, the per-file features in the group are analyzed to determine whether any of the per-file features can be in-source fragments. N ) data can be used for those per-file features of the group of corresponding features (ie, the same features identified in step 200), obtaining a representative MS2 (or MS N )data.

[0144] Representative MS2 (or MS N ) data can be the corresponding MS2 (or MS N ) data, or more usefully, standard MS2 (or MS N ) spectrum. Such a spectrum can be returned together with the identifier (for example, from mzCloud TMreturn), and can be configured to accurately simulate in-source fragmentation. Thus, in an embodiment, new functionality is added to the search engine (e.g., to mzCloud TM ), which returns the mass from a low-energy collision spectrum (e.g., HCD 10) as part of each identified hit.

[0145] In the event that identification 200 is unsuccessful and / or in the event that low energy collision data is not available, the MS2 (or MS N ) spectrum can alternatively be used as a representative MS2 (or MS N ). However, when using curated fragmentation data based on trust criteria, the allocation confidence is higher.

[0146] Then, representative MS2 (or MS N ) data with other per-file features in the group to determine whether any of the other per-file features are likely to be intra-source fragments of the per-file feature under consideration, for example by searching for a sequence in the group that has a similarity to a representative MS2 (or MS N )The quality of the source fragments in the data matches the quality of the per-file features.

[0147] Any per-file feature in the set that has such a matching quality is marked as potentially being in-source debris. Thus, for example, Figure 5D indicates that it has been temporarily assigned two user-defined base ions ([M+H] + or[M+Na] + ) or in-source fragment ions. Since the ion type of in-source fragments is usually unknown, a generic ion type assignment (e.g., [Me] + ).

[0148] For possible in-source fragments, the relationship between the possible in-source fragments and the precursor ions of the in-source fragments is also recorded in the graph. Figure 6 In , this potential relationship is indicated by the dashed line connection between nodes (the per-file feature labeled "P" is potentially a precursor ion for the in-source fragment "F").

[0149] Back again Figure 2In the next step (step 420), possible adduct relationships between the per-file features in the group are determined by finding the allowed mass shifts between the masses of the per-file features. A list of predetermined ion types (e.g., adducts, common neutral losses, simple multimers, and charge states) is used to generate expected mass shifts, which are applied to each group to find all possible relationships between the per-file features of the group. Thus, for example, in the case where two per-file features have a mass difference of 17 (=18-1), the two features may be protonated ions [M+H] + and its ammonia adduct [M+NH4] + ; The difference is 22 (= 23-1), and each file feature can be [M+H] + and [M+Na] + wait.

[0150] Any such possible relationships are recorded in the graph. Thus, for example, in Figure 6 , the solid line connections between nodes represent potential adduct relationships, where the possible adduct relationships (i.e., in this example, ) to mark the line.

[0151] Figure 6 A simple example of two per-file feature groups with various possible relationships labeled is shown. In particular, the first feature (on the left) has been labeled [M+H] + Type of identifying ion or in-source fragment [Me] + The second feature (on the right) is labeled as two different alkali ions, [M+H] + Or [M+NH4]+. Figure 6 In, marked as The solid lines represent possible adduct transitions, while the dashed lines represent possible in-source fragmentation relationships (the features labeled "P" are precursor ions for in-source fragmentation "F").

[0152] from Figure 6 As can be seen from the example of , the above process can lead to multiple possible, potentially conflicting relationships between the per-file features in the group. Figure 2 In step 430, these conflicts are resolved.

[0153] To do this, first, any conflicts between ion type assignments (cycles) and transitions (connections between nodes) are resolved. This process is illustrated in Figure 7 and involves (i) removing any transitions that conflict with the identified ions ( Fig. 7A ); and (ii) removing illegal debris transitions ( Figure 7B ).

[0154] Fig. 7A Again, it shows Figure 6 but with a second possible adduct transition (in addition to the transition Other than From Fig. 7A As can be seen, the source fragmentation mark [Me] for the first file feature has been removed + because this corresponds to the identified ion type [M+H] for the first per-file feature + Conflict. Then, the second possible adduct transition ( is removed because the transition requires [M+H-NH3] + Also with the identified ion type [M+H] for the first per-file feature + conflict.

[0155] Figure 7B An example of removing illegal fragment transitions is shown. Specifically, fragment transitions is removed, since it is not possible to generate a [M+H]+ fragment from a [M+Na]+ precursor (i.e., the fragmentation transition cannot "generate" a new element). Instead, the possible fragmentation transitions are kept Because [M+H]+ fragments can be generated from [M+NH4]+ precursors.

[0156] Figure 8 A simplified example per-file graph that can be produced by the above process is shown, where all remaining possible ion type assignments and transitions are shown by various loops and connections between nodes. It is still possible to combine several different interpretations of the per-file features. Figure 8 All possible clusters (each cluster corresponding to a single compound M) can be determined in the graph of .

[0157] Fig. 9 The results of this determination are shown. Fig. 9 , each possible cluster is shown as a shaded box. The leftmost box 431 represents a cluster that has no other competing interpretations (i.e., its per-file features do not appear in any other possible clusters). Therefore, this represents a successful identification of a cluster (at the per-file level) that does not require further processing. This cluster 431 is added to the main graph 510 (and later resolved in step 500 for any other competing interpretations from other sample files).

[0158] Each remaining cluster is a possible cluster that conflicts with at least one other possible cluster. These various conflicts are represented by Fig. 9 Next, these conflicts are resolved.

[0159] like Fig.10As shown, in an embodiment where the user has provided one or more alkali ion types (as described above), any possible clusters that do not include one or more per-file features assigned as potentially being one or more of the alkali ion types are rejected. (This step may be skipped in the event that the user does not provide any required ion-based features.) Thus, if Fig.10 As shown in A, clusters with alkali ions can be maintained. Fig.10 As shown in B, clusters with only identification ions can be rejected. Fig.10 As shown in C, clusters with only default ions can be rejected. Fig.10 As shown in D, clusters without any alkali ions can be rejected.

[0160] See again Fig. 9 In the example shown, the possible cluster labeled 432 is removed because only two transitions are found without any possible assignment (loop) of base adduct ions or identified adduct ions to any of the nodes within cluster 432. Therefore, cluster 432 is considered invalid and is not considered further.

[0161] Next, if Fig.11 As shown, after rejecting any invalid ion clusters, any remaining isolated clusters (i.e., individual per-file features that do not transition to any other per-file features) are considered. Fig.11 As shown in A, the assignment of any isolated clusters with a valid ion type assignment (i.e., base ion, identification ion, or default ion assignment) is retained. Fig.11 As shown in B, for isolated clusters flagged as potential in-source fragments, the ion type assignments are maintained only in the case of identified ions and common fragment ions (while the ion type assignments are removed in the case of base ions and default ions).

[0162] Thus, in an embodiment, step 430 includes (i) removing ion type assignments with mismatched charges; (ii) removing ion type assignments that conflict with any pre-identified ion types; (iii) removing any clusters that are missing one or more user-assigned alkali ion types; and (iv) removing any invalid fragment precursor relationships (e.g., where additional atoms need to be added to the fragment).

[0163] Fig.12 Shows that in the process Fig. 9 The graph obtained after the graph of . Fig.12 It can be seen that Fig. 9 Clusters 431 and 432 in have been removed (for the reasons given above). In addition, after removing cluster 432, Fig. 9 The isolated node 433 no longer conflicts with any other node, and is therefore considered resolved and has also been removed from Fig.12 is removed from the figure.

[0164] like Fig.12 As can be seen in , at this stage, there may still be multiple conflicting possible clusters. However, the graph has been parsed into two disconnected subgraphs, which can each be further processed separately because they do not conflict with each other.

[0165] Next, use the scoring system to resolve remaining conflicts. Fig.13 The process is shown, Fig.13 Shows Fig.12 The process of the leftmost subgraph in . Fig.12 A similar process will be performed separately for the other subgraph in .

[0166] During the scoring process, the largest possible cluster is first selected and assumed to be correct (“enabled”). Fig.13 As shown at A of , first enable the cluster labeled 434. Then disable any other clusters that conflict with this largest cluster. Next, enable the next largest possible cluster (not yet enabled or disabled), and so on, until the first interpretation (or "allocation" or "arrangement structure") is reached. Therefore, in Fig.13 In the example at A of , the bottom image represents this initial interpretation, where each file feature group is explained by two clusters 434 , 435 .

[0167] An assignment score for the interpretation is then given. To calculate the assignment score, each ion type assignment (cycle) is given a relative weight factor. Any suitable weight factor can be used. For example, the highest relative weight can be used for the most common adducts (e.g., [M+H] or [MH]), slightly lower weights can be used for common adducts (e.g., [M+Na] or [M+K]), and even lower weights can be used for the remaining adducts (e.g., uncommon adducts). The relative magnitude of the various weight factors can be configured as desired, for example to reflect the probability of a particular adduct occurring under the particular experimental conditions / sample chemistry used to obtain the data, and can be set by the user.

[0168] In addition, each transition is scored twice as high as the sum of its two ion weight factors. A factor of 2 is used here because transitions account for two nodes, whereas loop assignments account for only one node. Since the default assignment (loop) is usually the adduct with the highest weight, the algorithm will strongly prefer isolates that would otherwise be assigned as default adducts, which is undesirable.

[0169] Each cluster is then given a cluster score which is the sum of all cycle scores and transition scores of the cluster divided by the number of nodes in the cluster.

[0170] Finally, a partition score is calculated for the interpretation, which is the sum of all cluster scores for the clusters in the interpretation divided by the number of clusters. The partition score is designed to produce high scores for situations where there are many similar clusters (and to avoid interpretations that contain, for example, only one large cluster and many isolates).

[0171] Back again Fig.13 Next, test one or more other possible explanations. Fig.13 To do this, enable the cluster in the first step (i.e. Fig.13 A) and repeat the above process to generate competing explanations. So, for example, in Fig.13 The right image in B of FIG. 4 shows this competing interpretation, where each file feature group is now explained by three clusters 436, 437, 438. This interpretation is scored again using the same scoring system.

[0172] Again, another possible explanation is tested by enabling a different one of the clusters that was disabled in the first step, and repeating the above process to generate another competing explanation. Fig.13 In C of FIG. 4 , the right image represents this further competing interpretation, where each file feature group is now explained by two clusters 437, 439. This interpretation is again scored using the same scoring system.

[0173] This process can be repeated, for example, until all possible explanations are scored. The explanation with the highest assigned score can then be selected as the final explanation for output.

[0174] Therefore, possible clusters are recursively evaluated to arrive at the best possible assignment score, where the score is designed to reflect the properties of the sample being analyzed, e.g., under certain chromatographic conditions, most compounds should behave similarly and thus produce similar ion clusters.

[0175] Fig.14 An exemplary final interpretation is shown, in which the information from Fig. 9 This final interpretation is added to the main graph 510.

[0176] In the case where there are multiple sample files, a similar process is performed for each sample file, and one interpretation per sample file is added to the main graph 510 (and any duplications are removed). This may result in the main graph 510 having some conflicting interpretations (i.e., similar to Fig.12 ).

[0177] Therefore, return Figure 2 In step 500, any conflicting interpretations in the master graph 510 are resolved. Figure 12 to Figure 14) to perform the process using the same algorithm. This step ensures that information from all sample files is considered and a consistent allocation across files is achieved.

[0178] Each resulting subcluster represents a unique compound. Thus, in the end, compounds can be identified from each cluster in the final main graph ( Figure 2 Step 520).

[0179] While various specific embodiments have been described above, various alternative embodiments and additions are possible.

[0180] For example, a possible additional step in the processing pipeline may be a quantitative analysis of the results. For this purpose, it is necessary to know which features in the cluster can be used and which should be ignored (e.g., add up their areas for quantification). Usually, any fragments are ignored and only ions that are assigned as meaningful adducts are used for this. However, it is possible that fragments are used for quantification, for example when they are related to real adduct ions. For example, a common fragment of [M+NH4] is [M+H]. Therefore, an additional step can be applied to mark the real adduct ions for further processing, while any common fragments should be ignored.

[0181] The process is as Fig.15 As shown, Fig.15 As shown in A of , some fragments may be shared by multiple precursors. In this case, Fig.15 As shown in B of , if the relationship with one of the precursors is also a valid adduct transition, the relationship with any other precursor is rejected. This means that the fragment can be saved for further use while avoiding multiple uses. Fig.15 If there are some other adduct transitions from features for which no fragment-precursor relationship is detected for a fragment, those assignments are rejected and the fragment is assigned as a generic fragment ion, as shown in C of FIG.

[0182] The methods described herein are generally applicable, for example, to mass spectrometric analysis of small molecules, such as metabolic pathway analysis, degradation product analysis, forensic analysis, and the like.

[0183] Figure 16 shows example results for an artificial sample of eight compounds mixed in known ratios with the corresponding

[13] C labeled analogs. Due to the unique labeling, it was possible to identify the compound and evaluate the method along with all possible analogs.

[0184] like Fig.16A As shown in , the conventional technique clearly reports three additional compounds that show the expected isotope exchange curves, but have retention times close to one of the expected compounds. Considering the assigned molecular formulas, especially the formula differences between compounds at the same retention time, an incorrect interpretation is proposed.

[0185] like Fig. 16B As shown, the same data were analyzed by the method of compound combination according to the embodiment. In this case, all three additional compounds were correctly interpreted as in-source fragments and attached to their corresponding precursors.

[0186] The compound combination methods of various embodiments enable complete and consistent combination of multiple different forms of ions derived from a single compound within multiple samples. Unlike conventional techniques that are specifically based on expected mass shifts within MS1 ​​spectra, the embodiments additionally utilize MS2 fragmentation spectra and online identification tools for initial adduct assignment and for non-targeted in-source fragment detection. A set of chemical rules and heuristic rules are used to merge the results from these steps into unique ion clusters. This strategy significantly reduces the number of false positive identifications.

[0187] According to an embodiment, this is accomplished in particular by:

[0188] (i) A two-step partitioning mechanism is applied to merge potentially relevant features across multiple different sample files.

[0189] This enables consistent ion assignment across multiple different sample files.

[0190] (ii) Using the acquired MS2 data to pre-identify compounds to provide high-level

[0191] This avoids further ion mismatches of identifiable compounds.

[0192] Misallocation.

[0193] (iii) Using curated low-energy collision spectra of identified compounds in the data, and then searching for specific

[0194] This helps reduce the number of false positive unique compounds.

[0195] (iv) Using the acquired MS2 spectrum to search for matching in-source fragment candidates.

[0196] Reduce the number of false positive unique compounds when low energy collision spectra are not available.

[0197] (v) Apply a set of rules to the network of all possible ion relationships to remove those with low confidence

[0198] This reduces the chance of erroneous adduct assignment.

[0199] (vi) Apply custom scoring and evaluation mechanisms to the merged ion relationship network to address possible

[0200] This yields individual ion clusters corresponding to unique compounds.

[0201] Although the invention has been described with reference to various embodiments, it should be understood that various changes may be made without departing from the scope of the invention as set forth in the following claims.

Claims

1. A method for processing mass spectrometry data, wherein the mass spectrometry data comprises a plurality of MS 1 Mass spectrometry and multiple MS N (N≥2) mass spectra, each mass spectrum having a respective associated retention time, the method comprising: Detecting the multiple MS 1 a set of features in a mass spectrum, wherein each feature of the set has a corresponding mass, and wherein the features of the set have corresponding retention times; For each feature in the set of one or more features: (i) submitting the corresponding MS to a mass spectrometry search engine; N mass spectrum to obtain an identification result for the feature, and (ii) determining a candidate ion type for the feature based on a mass difference between the mass associated with the feature and an expected mass from the identification result; and then identifying one or more compounds based on the set of features and the candidate ion types.

2. The method of claim 1, wherein (ii) determining a candidate ion type for the feature comprises: A candidate adduct ion type for the feature is determined based on the mass difference between the mass of the feature and the expected mass from the identification result.

3. The method according to claim 1 or 2, wherein the mass spectrometry data comprises at least one sample file, each sample file corresponding to a respective chromatographic separation scan and comprising a plurality of MS 1 Mass spectrometry and multiple MS N mass spectrometry, and wherein the plurality of MS 1 The steps of feature group in mass spectra include: For each sample file: detecting a plurality of per-file features in the sample file, each per-file feature having a corresponding quality and a corresponding retention time; forming a plurality of features from the per-file feature, each feature having a corresponding quality and a corresponding hold time; and Feature groups are formed by dividing features having corresponding retention times.

4. The method of claim 3, wherein the step of detecting a plurality of per-file features in the sample files comprises: From the plurality of MS of the sample files 1 The mass spectrometer constructs multiple chromatograms, each chromatogram having a corresponding mass-to-charge ratio (m / z); determining a characteristic retention time for each chromatogram; dividing the chromatograms having corresponding characteristic retention times into one or more sets of chromatograms; as well as A deisotoping algorithm is applied to each chromatogram set to form a per-file feature set.

5. The method according to claim 4, wherein: The step of partitioning features having corresponding hold times includes partitioning features having hold times that are equal within a first tolerance; The step of dividing the chromatograms having corresponding characteristic retention times includes dividing the chromatograms having retention times that are equal within a second tolerance; and The second tolerance is smaller than the first tolerance.

6. The method according to claim 3, 4 or 5, wherein the mass spectrometry data comprises a plurality of sample files, and each of the plurality of features is formed by dividing a per-file feature, the per-file feature having a corresponding mass and a corresponding retention time.

7. The method of claim 6, wherein each feature group is formed by a corresponding per-file feature group in each sample file, and wherein the step of identifying one or more compounds based on the feature group comprises: For each per-file feature group in the corresponding sample file: (i) determining one or more per-file feature clusters, wherein each per-file feature cluster includes one or more per-file features of the set and may correspond to a corresponding compound; (ii) determining one or more arrangement structures of the per-file feature clusters for the per-file feature group, wherein each arrangement structure includes one or more non-conflicting per-file feature clusters; as well as (iii) selecting a preferred arrangement structure for each file feature group from the one or more arrangement structures of each file feature cluster; And then, based on the preferred arrangement structure for the plurality of sample files: (iv) determining one or more arrangement structures of feature clusters for the feature set, wherein each feature cluster includes one or more features of the feature set and may correspond to a corresponding compound, and wherein each arrangement structure includes one or more non-conflicting feature clusters; as well as (v) selecting a preferred arrangement for said feature group from said one or more arrangements of feature clusters; and then One or more compounds are identified based on the preferred arrangement of characteristic clusters.

8. A method for processing mass spectrometry data, wherein the mass spectrometry data comprises a plurality of sample files, each of which comprises a plurality of MS 1 Mass spectrometry and multiple MS N (N≥2) mass spectra, each mass spectrum having a respective associated retention time, the method comprising: For each sample file: detect the MS of the sample file 1 a plurality of per-file features in the mass spectrum, each per-file feature having a corresponding mass and a corresponding retention time; forming a plurality of features from the per-file features by partitioning the per-file features having corresponding qualities and corresponding retention times; Forming feature groups by dividing features with corresponding retention times, and forming a corresponding per-file feature group in each sample file; and then For each per-file feature group in the corresponding sample file: (i) determining one or more per-file feature clusters, wherein each per-file feature cluster includes one or more per-file features of the set and may correspond to a corresponding compound; (ii) determining one or more arrangement structures of the per-file feature clusters for the per-file feature group, wherein each arrangement structure includes one or more non-conflicting per-file feature clusters; as well as (iii) selecting a preferred arrangement structure for each file feature group from the one or more arrangement structures of each file feature cluster; And then, based on the preferred arrangement structure for the plurality of sample files: (iv) determining one or more arrangement structures of feature clusters for the feature set, wherein each feature cluster includes one or more features of the feature set and may correspond to a corresponding compound, and wherein each arrangement structure includes one or more non-conflicting feature clusters; as well as (v) selecting a preferred arrangement for said feature group from said one or more arrangements of feature clusters; and then One or more compounds are identified based on the preferred arrangement of characteristic clusters.

9. The method according to claim 7 or 8, wherein for each per-file feature group in the corresponding sample file, the step of determining one or more per-file feature clusters comprises: assigning one or more candidate ion types to each per-file feature of the set; determining one or more candidate relationships between the set of per-document features; as well as Any conflicts between the candidate ion types and the candidate relationships are resolved.

10. The method of claim 9, wherein each per-file feature has a corresponding charge, and wherein the step of assigning one or more candidate ion types to each per-file feature of the set comprises: assigning an identified ion type to any per-file feature of the set corresponding to a feature for which an identification result was obtained; and / or Each per-file feature of the set is assigned a user-defined base ion type or a default ion type based on the corresponding charge of the per-file features.

11. The method according to claim 9 or 10, wherein: The step of assigning one or more candidate ion types to each per-file feature of the set comprises: assigning an in-source fragment ion type to any per-file feature of the set whose mass corresponds to the mass of an expected in-source fragment of another per-file feature in the set; and / or The step of determining one or more candidate relationships between the per-file features of the group includes: when the quality of the per-file feature corresponds to the quality of the expected in-source fragmentation of another per-file feature of the group, determining the in-source fragmentation relationship between the per-file feature of the group and the other per-file feature of the group.

12. The method of claim 11, further comprising obtaining the quality of expected in-source fragmentation per file feature by: From the MS corresponding to each file feature N Mass spectrometry determines the masses of the expected in-source fragments for the per-file features.

13. The method according to claim 11 or 12, further comprising obtaining the quality of expected in-source fragmentation per file feature by the following steps: The masses of one or more expected in-source fragments of a feature are provided as part of the identification result for the feature.

14. A method for processing mass spectrometry data, wherein the mass spectrometry data comprises a plurality of MS 1 Mass spectrometry and multiple MS N (N≥2) mass spectra, each mass spectrum having a respective associated retention time, the method comprising: Detecting the multiple MS 1 a set of features in a mass spectrum, wherein each feature of the set has a corresponding mass, and wherein the features of the set have corresponding retention times; For each feature in the set of one or more features: (i) submitting the corresponding MS to a mass spectrometry search engine; N mass spectrum to obtain an identification result for the feature, and (ii) providing the mass of one or more expected in-source fragments of the feature as part of the identification result; assigning an in-source fragment ion type to any feature of the set whose mass corresponds to the mass of an expected in-source fragment of another feature in the set; and One or more compounds are identified based on the set of features and the in-source fragment ion types.

15. The method of claim 13 or 14, wherein the step of: N A mass spectrometer determines the mass that is provided as part of the identification result.

16. The method according to any one of claims 9 to 15, wherein the step of determining one or more candidate relationships between the set of per-file features comprises: One or more candidate adduct relationships between the set of per-file features are determined based on the allowed mass shifts between the set of per-file features.

17. The method according to any one of claims 7 to 16, wherein the step of selecting a preferred arrangement of clusters from the one or more arrangements of clusters comprises: Determine the score for each arrangement of clusters; as well as The arrangement with the highest score is selected.

18. The method of claim 17, wherein the step of determining a score for each arrangement of clusters comprises: Determining a cluster score for each cluster in the arrangement structure by: (i) assigning a weight factor to each candidate ion type assignment for the cluster, (ii) assigning a relationship score to each candidate relationship for the cluster, and (iii) calculating the cluster score for the cluster by dividing the sum of the weight factor and the relationship score by the number of features in the cluster or the number of features per file; as well as The score for each arrangement is determined by dividing the sum of the cluster scores for the arrangement by the number of clusters in the arrangement.

19. A mass spectrometry analysis method, comprising: The samples were analyzed to obtain multiple MS 1 Mass spectrometry and multiple MS N mass spectral data of mass spectra, each mass spectrum having a corresponding associated retention time; as well as The mass spectrometry data is processed using a method according to any one of the preceding claims.

20. A non-transitory computer readable storage medium storing computer software code which, when executed on a processor, performs the method according to any one of the preceding claims.

21. A control system for an analytical instrument, the control system being configured to cause the analytical instrument to perform the method according to any one of claims 1 to 19.

22. An analytical instrument comprising a control system according to claim 21.

Images