Mass analysis data processing method and mass analysis device

Abstract

The invention relates to a mass analysis data processing method and a mass analyzer. The device is provided with: a measurement unit (1) that acquires data by performing mass analysis on a sample; a peak information acquisition unit (34) that performs peak detection on an m / z spectrum based on the data and collects peak information including m / z values of a plurality of detected peaks; an approximate mass calculation unit (35) that calculates an approximate mass by multiplying the m / z value of each peak included in the peak information by each of a plurality of prices within a predetermined expected price range; a rank selection unit (36, 37) for determining, as the likelihood of the budget quality, the frequency of the rank for each budget quality or for each predetermined quality width for the plurality of budget qualities, and selecting the budget quality or rank that is estimated to be highly reliable on the basis of the likelihood; and an estimated mass calculation unit (38) that calculates, for the one or more estimated masses or the estimated masses included in the selected one or more estimated masses or ranks, the estimated mass of the compound corresponding to the estimated masses on the basis of the peak information and the valence number of the corresponding peaks when the respective estimated masses are calculated.

Description

Technical Field

[0001] The present invention relates to a data processing method for processing data collected by a quality analysis device, and a quality analysis device using the data processing method. Background Technology

[0002] When analyzing polymeric compounds such as antibodies and proteins using mass spectrometry employing electrospray ionization (ESI) plasma ionization, multivalent ions with a wide range of valences are detected. Furthermore, multivalent ions originating from adducts formed by the addition of various substances to the target polymeric compound are frequently detected. Moreover, when using mass spectrometry with high mass resolution, a peak of a particular valence is further detected as multiple distinct peaks corresponding to different isotopic ratios, resulting in numerous isotopic envelopes in the mass spectrum. Therefore, even if the sample contains only one compound, the mass spectrum obtained through mass analysis exhibits a highly complex pattern. Consequently, it is difficult for users to visually identify peaks originating from the target compound with known valences and isotopic ratios from such mass spectra and to infer the mass based on the mass-to-charge ratio (m / z) of those peaks.

[0003] Therefore, in the analysis of such mass spectrometry data, a data processing method called charge deconvolution is typically performed to attribute multiple peaks selected from the mass spectrometer originating from each molecule to the mass of that molecule (see Non-Patent Document 1). Various algorithms are known for this charge deconvolution; the algorithm described in Non-Patent Document 2 uses a model function, while the algorithm described in Non-Patent Document 3 utilizes Bayesian inference.

[0004] Existing technical documents

[0005] Non-patent literature

[0006] Non-patent literature 1: Iain DGCampuzano et al., 11, “Natural and Denaturing MS Protein Deconvolution for Biopharma: Monoclonal Antibodies and Antibody-Drug-Conjugates to Polydisperse Membrane Proteins and Beyond,” Analytical Chemistry, 2019, Vol. 91, No. 15, pp. 9472-9480, (Internet)<URL:https: / / doi.org / 10.1021 / acs.analchem.9b00062> )

[0007] Non-patent literature 2: Marshall Bern et al., 12, “Parsimonious Charge Deconvolution for Native Mass Spectrometry”, Journal of Proteome Research, 2018, Vol. 17, No. 3, pp. 1216-1226 (Internet)<URL:https: / / doi.org / 10.1021 / acs.jproteome.7b00839> )

[0008] Non-patent literature 3: Michael T. Marty et al., “Bayesian Deconvolution of Mass and Ion Mobility Spectra: From Binary Interactions to Polydisperse Ensembles,” Analytical Chemistry, 2015, Vol. 87, pp. 4370-4376 (Internet)<URL:https: / / doi.org / 10.1021 / acs.analchem.5b00140> )

[0009] Non-Patent Literature 4: “UniDec: Universal Deconvolution of Mass and Ion Mobility Spectra”, MARTY LAB, [Online], [Searched May 9, 2024], Internet<URL:https: / / martylab.arizona.edu / software> Summary of the Invention

[0010] The technical problem that the invention aims to solve

[0011] In charge deconvolution methods described above, which utilize various computational techniques, not only is the average mass of detected molecules or adducts calculated, but also charge deconvolution methods used to infer the mass of single isotopes with small ratios and weak signals that do not appear in the mass spectrum, or to infer isotopic envelopes that reflect mass resolution. However, these conventional charge deconvolution methods sometimes suffer from artifacts that are not actually present in the results (e.g., peak-shaped artifacts). Furthermore, the computation is generally complex, leading to problems such as the need for large amounts of memory or unpredictable convergence times. Moreover, depending on the charge deconvolution algorithm, it can be difficult to optimize the parameters to obtain the desired results.

[0012] This invention was made to solve such technical problems, and its main objective is to provide a quality analysis data processing method and quality analysis device that can avoid the generation of artifacts in charge deconvolution and obtain highly reliable processing results.

[0013] Solution to the above technical problems

[0014] One aspect of the quality analysis data processing method of the present invention is a data processing method for processing data obtained by performing quality analysis on a sample, comprising:

[0015] The peak information acquisition step involves peak detection on the m / z spectrum based on the acquired data, and collecting peak information containing the m / z values ​​of multiple detected peaks.

[0016] The preliminary quality calculation step involves multiplying the m / z value of each peak contained in the peak information by multiple values ​​within a predetermined range of assumed values ​​to calculate the preliminary quality.

[0017] In the grade selection step, for the multiple estimated qualities calculated in the estimated quality calculation step, the frequency of each estimated quality or grade of each specified quality width is calculated as the likelihood of the estimated quality. Based on the likelihood, the estimated quality or grade that is inferred to have high reliability is selected.

[0018] The inferred quality calculation step involves calculating the inferred quality of a compound corresponding to one or more estimated qualities or grades selected in the grade selection step, based on the peak information and valence of the corresponding peaks when each estimated quality was calculated in the estimated quality calculation step.

[0019] Furthermore, one embodiment of the quality analysis apparatus of the present invention includes:

[0020] The measurement department acquires data by performing quality analysis on the samples;

[0021] The peak information acquisition unit performs peak detection on the m / z spectrum based on the data acquired by the measurement unit and collects peak information including the m / z values ​​of multiple detected peaks;

[0022] The preliminary quality calculation unit calculates the preliminary quality by multiplying the m / z value of each peak contained in the peak information by multiple values ​​within a predetermined range of assumed values.

[0023] The grade selection unit calculates the frequency of each estimated quality or grade within each specified quality width for multiple estimated qualities calculated by the estimated quality calculation unit as the likelihood of the estimated quality, and selects estimated qualities or grades that are inferred to have high reliability based on the likelihood.

[0024] The inferred quality calculation unit calculates the inferred quality of the compound corresponding to the estimated quality for one or more estimated qualities or estimated qualities included in the grade selected by the grade selection unit, based on the peak information and valence of the corresponding peak when each estimated quality is calculated by the estimated quality calculation unit.

[0025] The m / z spectrum mentioned here is a mass spectrum, but the horizontal axis is the m / z spectrum (profile spectrum) instead of mass.

[0026] Invention Effects

[0027] In the mass analysis data processing method and mass analysis apparatus of the present invention, the mass is calculated based on the principle that the mass value is obtained by multiplying the m / z value corresponding to the ion peak observed in the m / z spectrum by the correct valence of the ion. Therefore, in principle, no artifacts are generated, and the accuracy of the calculated mass value of the target compound in the sample can be improved, resulting in highly reliable analytical results. Furthermore, the main processing involves simple repetitive calculations without complex computations, thus reducing memory consumption during processing and lessening the load on computers and other equipment. In addition, it is easy to predict the progress until the result is obtained and the time required until the end of processing, which helps to improve the efficiency of the analysis operation. Attached Figure Description

[0028] Figure 1 This is a schematic modular configuration diagram of a quality analysis apparatus according to an embodiment of the present invention.

[0029] Figure 2 This is a flowchart illustrating the steps of charge deconvolution-based data processing in the quality analysis apparatus of this embodiment.

[0030] Figure 3 This is a diagram showing an example of the m / z spectrum as the object of analysis.

[0031] Figure 4 This is an explanatory diagram illustrating the definitions of the left and right ends of the peaks detected in the m / z spectrum.

[0032] Figure 5 This is a graph illustrating an example of the relationship between quality level and score obtained during the data processing in this embodiment.

[0033] Figure 6 This is a simplified diagram illustrating the method for calculating the frequency (histogram) of estimated quality in the data processing of this embodiment.

[0034] Figure 7 This is a graph showing a comparison between the mass spectra obtained using the data processing of this embodiment and those obtained using existing software. Detailed Implementation

[0035] In this specification, m / z spectrum refers to mass spectrum (profile spectrum) with m / z as the horizontal axis, and mass spectrum refers to profile spectrum with mass as the horizontal axis.

[0036] Furthermore, in the quality analysis data processing method and quality analysis apparatus of the above-described scheme of the present invention, a typical quality analysis method is the quality analysis method of ionization, such as ESI, which easily generates multivalent ions, especially multivalent ions with a wide range of valences.

[0037] Furthermore, there are no particular limitations on the method of separating ions based on m / z; for example, suitable devices such as quadrupole mass filters, time-of-flight mass separators, and ion cyclotron resonance mass separators can be used. Of course, for high-precision determination of the mass value, a mass separator with high mass resolution is desirable. In this regard, time-of-flight mass separators, especially multiple-orbit time-of-flight mass separators or multiple-reflection time-of-flight mass separators, or ion cyclotron resonance mass separators are useful.

[0038] Hereinafter, an embodiment of the quality analysis data processing method of the present invention and the quality analysis apparatus using the data processing method will be described with reference to the accompanying drawings.

[0039] Figure 1 This is a schematic diagram of the modular structure of the quality analysis device in this embodiment.

[0040] exist Figure 1 In this device, the measurement unit 1 is a mass analysis apparatus using an ESI source, including an ESI unit 10, a mass separation unit 11, and an ion detection unit 12. The mass separation unit 11 is expected to have high quality resolution; for example, a time-of-flight mass separator configured as a reflection type, multiple reflection type, or multiple surround type can be used. The analysis control unit 2 controls the measurement unit 1, and the data processing unit 3 processes the detection signal obtained through the measurement unit 1.

[0041] The data processing unit 3 includes an MS analysis data collection unit 30, a data storage unit 31, an analysis condition setting unit 32, a valence range determination unit 33, a peak detection unit 34, a preliminary quality calculation unit 35, a likelihood calculation unit 36, a quality grade selection unit 37, and a mass spectrometry generation unit 38 as functional modules. The data processing unit 3 is connected to an input unit 4 for users to input specified parameters and a display unit 5 for displaying data processing results.

[0042] At least a portion of the analysis and control unit 2 and the data processing unit 3 are capable of using a personal computer or a higher-performance computer as hardware resources, and performing their respective functions on the computer by working on the computer through software (computer programs) installed on the computer.

[0043] In the quality analysis apparatus of this embodiment, quality analysis data for the sample is acquired as follows.

[0044] If a sample containing one or more compounds is introduced into the ESI section 10, the ESI section 10 ionizes the compound molecules contained in the sample. It is well known that ionization in the ESI generates multivalent ions with a wide range of valences. Furthermore, adduct ions, formed by the addition of substances such as alkali metals contained in the sample to the target compound in the sample, are sometimes generated. The generated ions are then introduced into the mass separation section 11 via an ion guide (not shown), where they are separated according to their m / z. For example, in the case where the mass separation section 11 is a time-of-flight mass separator, the various ions introduced into the mass separation section 11 at approximately the same time are spatially separated according to their m / z during their flight in the flight space, arriving sequentially at the ion detection section 12 from the ions with the smallest m / z. The ion detection section 12 outputs a detection signal with an intensity corresponding to the number (quantity) of arriving ions.

[0045] In the data processing unit 3, the MS analysis data collection unit 30 digitizes the detection signal received from the measurement unit 1. Then, the MS analysis data collection unit 30 stores the m / z spectrum data, which shows the relationship between m / z and intensity, in the data storage unit 31. That is, this m / z spectrum data is the original profile data. However, data after appropriate noise processing, etc., has been applied to the original profile data may also be stored in the data storage unit 31 as the m / z spectrum data.

[0046] With the m / z spectrum data obtained from the mass analysis of the sample stored in the data storage unit 31, data processing including charge deconvolution is performed. Figure 2 This is a flowchart illustrating an example of the steps involved in the data processing.

[0047] In the following description, as a specific example, we illustrate the case where a monoclonal antibody (NISTmAb, molecular weight approximately 145,000 Da) deglycosylated by digestive enzymes is used as the target compound, and data processing is performed on the m / z spectral data obtained by mass analysis of the full length of the compound. Figure 3 This is an example of the m / z spectrum obtained for this compound. Furthermore, this data was obtained using a Shimadzu Q-TOF mass spectrometer, the "LCMS-9050". Figure 3 It can be seen that multivalent ions with a wide range of valences are generated during mass analysis, so multiple peaks of multivalent ions are mainly observed in the m / z spectrum.

[0048] If data processing begins, the analysis condition setting unit 32 receives the expected mass range M predicted for the target compound molecule, input by the user through the input unit 4. min ~M max And the hypothetical m / z range (m / z) used for charge deconvolution in the m / z spectrum as the object of analysis. min~(m / z) max As a parameter (step S1). Typically, the envisioned mass range can be predicted based on prior information about the target compound. Furthermore, the envisioned m / z range can be, for example, based on... Figure 3 The range of peaks with significant observable heights in the m / z spectrum is used to determine this. However, either or both of the assumed mass range and the assumed m / z range may not use user-inputted values, but rather values ​​determined as default values. If both default values ​​are used, step S1 is skipped.

[0049] The price range determination unit 33 determines the expected price range Z based on the expected quality range and expected m / z range set in step S1 using the following formula (1). min ~Z max (Step S2). Additionally, assuming both the assumed mass range and the assumed m / z range are default values, the assumed valence range Z... min ~Z max You can set it to the default value as well.

[0050] Z min =M min / {(m / z) max -m p}, Z max =M max / {(m / z) min -m p}…(1)

[0051] Where, m p The mass of the proton is 1.00728 Da.

[0052] Peak detection unit 34 obtains m / z spectrum data within the assumed m / z range set (or determined by default) in step S1 from data storage unit 31, and performs peak detection according to a prescribed peak detection algorithm. Then, it assigns a number to each detected peak to determine its position (step S3). Here, the leftmost peak in the m / z spectrum is designated as peak 1, and peaks are sequentially assigned numbers increasing by 1 towards the right (i.e., the direction in which the m / z value increases).

[0053] m / z spectral data is a collection of data points grouped by m / z values ​​and intensity values. A peak is composed of multiple data points along the m / z axis. Therefore, the peak detection unit 34 collects the m / z values ​​and intensity values ​​of the data points belonging to each peak and temporarily stores them as peak information in the data storage unit 31 (step S4). Here, "belonging to a peak" means, for example, that it can be included in the m / z range from the start point to the end point of the peak, or in the m / z range between the left and right ends of the peak as described later.

[0054] Furthermore, for each detected peak, the peak detection unit 34 calculates the m / z value corresponding to the data point on the left and the m / z value corresponding to the data point on the right (step S5). Here, the m / z value at the left end and the m / z value at the right end of the j-th peak from the left are respectively labeled as (m / z). j_L and (m / z) j_R It has an interval (m / z) enclosed by these m / z values ​​on the left and right. j_L ~(m / z) j_R The m / z spectral data of the m / z values ​​within the range are treated as data belonging to the same peak.

[0055] The terms "data point equivalent to the left end" and "data point equivalent to the right end" as used here refer to any one of the following definitions. Figure 4 This is an explanatory diagram illustrating the definitions of the left and right ends of a peak.

[0056] As an example, the left and right ends can be used when calculating the half-width (fwhm) of each peak, representing the left and right ends respectively. According to this definition, in Figure 4 Select the data point located at the position indicated by the black triangle. Alternatively, the centroid of the peak can be used as the center, and the range of twice the half-width (2fwhm) can be used as the peak region, along with the left and right ends of this region. According to this definition, in... Figure 4 In the selection process, choose the data points located at the positions indicated by the white triangles. Alternatively, other definitions can be used to determine the left and right ends of each peak.

[0057] Furthermore, it can be configured so that the user can choose which of the multiple definitions described above to use. Alternatively, it can be configured so that if the user specifies the quality resolution R from the input unit 4, the peak region is determined by referring to that value, and the left and right ends are determined based on that peak region. Specifically, for example, not only extracting Figure 4 The data points (× marked) in a peak region shown are further extracted into a set of continuous data points with intensity above the baseline, in which the m / z value of the peak apex or the m / z value of the centroid are calculated. Then, based on this m / z value and the quality resolution R, Δ(m / z) is calculated using equation (2) described later, and this value is used instead of Figure 4 The peak region can be determined by the half-width value fwhm shown.

[0058] Next, the preliminary quality calculation unit 35 calculates the m / z value (m / z) for the left end of all peaks. j_L and the m / z value on the right (m / z) j_R Multiply by the expected value range Z determined in step S2 respectively. min ~Z max All values ​​Z i The estimated mass M obtainedij_L M ij_R (Step S6). Furthermore, when processing the estimated quality of a certain peak, in order to easily determine the valence Z used to calculate that estimated quality... i The peak number j is used to generate an index that records the correspondence between the estimated quality, price, and peak number, and this index is temporarily stored in the data storage unit 31.

[0059] Next, the likelihood calculation unit 36 ​​uses the estimated mass M calculated in step S6. ij_L M ij_R This generates a histogram with quality grade on the horizontal axis and frequency (occurrence frequency) on the vertical axis. Specifically, a predetermined quality width ΔM is set on the horizontal axis of the histogram. edge Quality grade M edge_1 M edge_2 ..., find the approximate quality M that contains each peak. ij_L M ij_R Quality grade M edge_j The frequency of this quality level is counted incrementally. Furthermore, the m / z value (m / z) sandwiched on the left end... j_L Compared to the m / z value on the right (m / z) j_R Similarly, the m / z values ​​of all data points belonging to the same peak are also multiplied by the assumed value range Z. min ~Z max All values ​​Z i After the estimated quality is obtained, it is reflected in the histogram (step S7).

[0060] Here, the quality grade M on the horizontal axis edge_t (Where t = 1, 2, ...) can be within the envisioned mass range set in step S1, with a mass width ΔM of a pre-determined internal step size. edge To set it. For example, the mass width ΔM edge It can be set as the lower limit M of the envisioned quality range set in step S1. min Or upper limit M max Values ​​on the order of 10 ppm to 100 ppm. This mass width ΔM edge Preferably, the mass width is appropriate, corresponding to the mass resolution during measurement or corresponding to the mass resolution (or its range) specified in the device, but larger than that mass resolution. This is because, if the mass width ΔM is made larger relative to the mass resolution... edge If the range is too narrow, the estimated quality will be scattered across different quality grades, making it difficult to generate quality grades with high frequency.

[0061] Figure 6 This is a very simplified illustration of how the histogram is calculated in step S7. Now, let's consider the hypothetical price range Z. min~Z max The price Z within i Considering the three peaks with valences of i-1, i, and i+1, and the three peaks with peak numbers of j-1, j, and j+1, consider the following: Figure 6 As shown, this combination yields 6 sets of estimated mass M. ij_L M ij_R In relation to the estimated quality M ij_L and approximate quality M ij_R The corresponding quality levels (in) Figure 6 The frequency is counted incrementally within the mass level reached by the dashed line extending vertically upwards from the black circle mark ●. Therefore, for example, there exist two approximate mass levels M. i(j-1)_L M (i-1)j_L Quality grade M edge_1 The frequency is 2. On the other hand, only the approximate quality M exists. (i-1)j_R Quality grade M edge_3 The frequency is 1. Furthermore, in a region located with a set of estimated mass M... ij_L M ij_R Within the corresponding quality levels, data points belonging to the same peak are counted incrementally as corresponding data points. Therefore, for example, in the estimated quality M... (i-1)j_L With estimated quality M (i-1)j_R Quality grade M edge_2 The frequency is incremented by 2. This yields a histogram showing the frequency of occurrence of the approximate quality corresponding to each peak.

[0062] The vertical axis of such a histogram, i.e., the frequency, can be interpreted as the likelihood of the estimated mass of the compound molecules in the sample, inferred from the m / z spectrum, corresponding to the quality grade. Therefore, in this data processing method, this frequency, i.e., the likelihood, is treated as a "score." The score is an integer value greater than or equal to 0. Once the histogram is complete, it is necessary to easily determine the quality grade M. edge_t estimated quality M k (M ij_L ≤M k ≤M ij_R An index showing the correspondence between quality levels and estimated quality can be generated in advance and temporarily stored in the data storage unit 31.

[0063] Quality grade selection section 37 selects one or more quality grades M that score higher in the above histogram. edge_u (Step S8). As a method for selecting quality levels, for example, it can be set so that a threshold is specified by the user in step S1, thereby selecting all quality levels with scores above that threshold. Alternatively, it can be set to select a predetermined number of quality levels in descending order of scores, regardless of the score value.

[0064] Figure 5 This is an example of a histogram generated based on measured m / z spectra. In this example, the score threshold is set to S. th Select the quality grade indicated by the prominent black triangle mark.

[0065] Next, the mass spectrometry generation unit 38, based on the mass level M selected in step S8, performs the following steps: edge_u estimated quality M k The corresponding information is used to generate a newly defined horizontal axis M. cand_m Vertical axis I cand_m The mass spectrometer was used to calculate the estimated mass M. k The more accurate mass (inferred mass) of the corresponding compound (step S9).

[0066] Specifically, by referring to the index showing the correspondence between the quality level and the estimated quality generated in step S7, the selected quality level M can be easily determined. edge_u Multiple estimated quality M k Furthermore, by referring to the index that records the correspondence between the estimated quality, valence, and peak number generated in step S6, the estimated quality M can be easily determined and calculated. k The corresponding valence and peak number. If the peak number can be determined, the m / z value and intensity value corresponding to that peak obtained in step S4 can be easily calculated. However, even without referring to such an index, the quality level M selected in step S8 can be determined by sequentially tracing the calculation results. edge_u The m / z value and intensity value of the corresponding peak data points.

[0067] The positions M of each point on the horizontal axis (mass axis) of the mass spectrometer cand_1 M cand_2 ...able to achieve the selected quality level M edge_u Within a specified margin range before and after the center, for example, based on the mass resolution R of the mass analyzer used and the number of data points n constituting a peak. p The step size w is determined by the interval. That is, the quality resolution R is...

[0068] R=(m / z) / Δ(m / z)=M / ΔM…(2)

[0069] Therefore, the step size w is set to

[0070] w = ΔM / n p =M / (n p R)

[0071] That's all.

[0072] During peak detection in step S3, the quality resolution R can be pre-calculated based on representative peaks among the detected peaks, and the number of data points constituting the peak can be pre-determined and set as the number of data points n. p Alternatively, the user can input the quality resolution and the number of peak data points as one of the parameters for data processing in step S1, and then use these parameters for processing in step S9.

[0073] If the mass spectrometry generation unit 38 determines the step size of the horizontal axis of the mass spectrum as described above, then M will be satisfied. cand_m ≤M k <M cand_(m+1) estimated quality M k All as M cand_m To handle it. Therefore, as with a certain mass M cand_m The corresponding value on the vertical axis is the intensity I. cand_m Is it satisfying M? cand_m ≤M k <M cand_(m+1) Total estimated mass M k The sum of the intensities Σ k I k This is the sum of the intensity values ​​of all data points belonging to the peak corresponding to the quality level selected in step S8. In other words, this process is equivalent to performing a merging process for each quality level selected in step S8, which merges the intensity values ​​of multiple data points corresponding to that quality level. Thus, by applying the mass values ​​according to the mass step size w, i.e., in M... cand_1 M cand_2 By calculating the intensity values ​​at each position of the mass spectrum, a mass spectrum can be generated.

[0074] Furthermore, if a mass spectrum is obtained, the mass spectrometry generation unit 38 acquires the mass value corresponding to the position of the peak apex in the mass spectrum as the inferred mass of the target compound. Then, the mass spectrometry generation unit 38 displays the generated mass spectrum on the screen of the display unit 5 according to the user's instructions via the input unit 4.

[0075] Figure 7 (A) is a quality class M. edge_u estimated quality M k and the intensity I of the corresponding data points k The mass step size w, set according to the mass resolution of the mass analysis device used in the measurement, is plotted as M. cand_m and intensity I cand_m Mass spectrometry was generated. On the other hand, Figure 7 (B) shows the relationship with Figure 7For case (A), the same m / z spectral data were set with the same assumed mass range and assumed m / z range, and the mass spectrometry of the charge deconvolution results was performed using the existing software "UniDec" (refer to non-patent literature 4).

[0076] Compare Figure 7 As shown in (A) and (B), the mass spectrum generated by the above data processing method is depicted with higher peak resolution, i.e., higher precision, compared to the mass spectrum generated by "UniDec". Therefore, it can be concluded that by using this data processing method, high-resolution mass spectra can be generated, and the mass of the target compound can be inferred with high precision.

[0077] Furthermore, when the sample contains multiple compounds, multiple mutually separated mass levels can be selected corresponding to each compound. In this case, a mass spectrum can be generated for each mass level.

[0078] As described above, in the mass analysis apparatus of this embodiment, high-precision mass spectra can be obtained by performing characteristic data processing on the m / z spectra of peaks and isotopic envelopes of multivalent ions or adduct ions with a wide range of valences. Furthermore, highly accurate mass values ​​of the target compound can be obtained from this mass spectra.

[0079] The quality analysis data processing method and quality analysis apparatus of the above embodiments can be modified in various ways, in addition to the matters already described. For example, in the description of the data processing steps above, various indexes are generated during the processing, but as mentioned above, this is to shorten the processing time by referring to the indexes, and it is obviously possible to perform the same processing without generating indexes.

[0080] Alternatively, the data processing can be performed sequentially from steps S1 to S9, but for example, the data generated in step S7 can be processed sequentially. Figure 5 The histogram shown is temporarily displayed on the screen of display unit 5, allowing the user to visually confirm the histogram and then set a score threshold, enabling the user to participate midway through the processing. Furthermore, it is also possible to perform peak detection before... Figure 3 The m / z spectrum shown is temporarily displayed on the screen of display unit 5, allowing the user to visually confirm the m / z spectrum and then instruct to continue analysis.

[0081] As a result, users can identify situations earlier, such as when the overall score is too low to be suitable for selecting a quality level, and thus be able to re-evaluate the parameters or discover problems with the measurement itself.

[0082] Furthermore, in the above embodiments, such as Figure 6As shown, the number of estimated qualities included in the quality level is counted, and the frequency corresponding to each quality level is calculated as the score. However, when the estimated quality is a discrete value with a relatively large step size, the frequency of occurrence of each value of that estimated quality, rather than the quality level, can also be calculated as the score.

[0083] Furthermore, the above-described embodiments or various modifications are merely examples of the present invention, and appropriate modifications, alterations, additions, etc., made within the scope of the spirit of the present invention are naturally included within the scope of the claims of this application.

[0084] [Various options]

[0085] The exemplary embodiments described above are specific examples of the following solutions, which will be obvious to those skilled in the art.

[0086] (Item 1) One aspect of the quality analysis data processing method of the present invention is a data processing method for processing data obtained by performing quality analysis on a sample, comprising:

[0087] The peak information acquisition step involves peak detection on the m / z spectrum based on the acquired data, and collecting peak information containing the m / z values ​​of multiple detected peaks.

[0088] The preliminary quality calculation step involves multiplying the m / z value of each peak contained in the peak information by multiple values ​​within a predetermined range of assumed values ​​to calculate the preliminary quality.

[0089] In the grade selection step, for the multiple estimated qualities calculated in the estimated quality calculation step, the frequency of each estimated quality or grade of each specified quality width is calculated as the likelihood of the estimated quality. Based on the likelihood, the estimated quality or grade that is inferred to have high reliability is selected.

[0090] The inferred quality calculation step involves calculating the inferred quality of a compound corresponding to one or more estimated qualities or grades selected in the grade selection step, based on the peak information and valence of the corresponding peaks when each estimated quality was calculated in the estimated quality calculation step.

[0091] (Item 6) One aspect of the quality analysis apparatus of the present invention comprises:

[0092] The measurement department acquires data by performing quality analysis on the samples;

[0093] The peak information acquisition unit performs peak detection on the m / z spectrum based on the data acquired by the measurement unit and collects peak information including the m / z values ​​of multiple detected peaks;

[0094] The preliminary quality calculation unit calculates the preliminary quality by multiplying the m / z value of each peak contained in the peak information by multiple values ​​within a predetermined range of assumed values.

[0095] The grade selection unit calculates the frequency of each estimated quality or grade within each specified quality width for multiple estimated qualities calculated by the estimated quality calculation unit as the likelihood of the estimated quality, and selects estimated qualities or grades that are inferred to have high reliability based on the likelihood.

[0096] The inferred quality calculation unit calculates the inferred quality of the compound corresponding to the estimated quality for one or more estimated qualities or estimated qualities included in the grade selected by the grade selection unit, based on the peak information and valence of the corresponding peak when each estimated quality is calculated by the estimated quality calculation unit.

[0097] Based on the mass analysis data processing method described in Section 1 and the mass analysis apparatus described in Section 6, the mass is calculated by multiplying the m / z value corresponding to the ion peak observed in the m / z spectrum by the correct valence of the ion. Therefore, in principle, no artifacts are produced. This improves the accuracy of calculating the mass value of the target compound in the sample, resulting in highly reliable analytical results. Furthermore, the main processing involves simple repetitive calculations without complex computations, thus minimizing memory consumption and reducing the load on computers. Additionally, the progress towards obtaining results and the time required to complete the processing are easily predictable, which improves the efficiency of the analysis operation.

[0098] (Item 2) In the mass analysis data processing method described in Item 1, it can be set such that the peak information includes the intensity value of the peak in addition to the m / z value of the peak, and the inferred mass calculation step includes a spectrum generation step, which generates a mass spectrum with mass as the horizontal axis and intensity value as the vertical axis based on the peak information and valence of the peak corresponding to the estimated mass.

[0099] (Item 7) In the mass analysis apparatus described in Item 6, the peak information can be configured to include the intensity value of the peak in addition to the m / z value of the peak, and the inferred mass calculation unit includes a spectrum generation unit that generates a mass spectrum with mass on the horizontal axis and intensity value on the vertical axis based on the peak information and valence of the peak corresponding to the estimated mass.

[0100] In addition, in the mass analysis data processing method described in item 2 and the mass analysis apparatus described in item 7, the horizontal axis of the mass spectrometer can be set to the mass width of the step size corresponding to the mass resolution of the mass analysis apparatus used in the measurement.

[0101] Based on the mass analysis data processing method described in item 2 and the mass analysis apparatus described in item 7, it is possible to determine the mass of the target compound in the sample based on the peak positions and other parameters in the generated high-precision mass spectrum. Therefore, in addition to providing the user with the mass of the target compound, information reflecting the isotopic distribution can also be provided.

[0102] (Item 3) In the mass analysis data processing method described in Item 2, it can be set such that, in the spectrum generation step, the mass spectrum is generated by merging the data with a mass width corresponding to the mass resolution of the peaks in the m / z spectrum.

[0103] (Item 8) In the mass analysis apparatus described in Item 7, the spectrum generation unit can be configured to generate a mass spectrum by merging data with a mass width corresponding to the mass resolution of the peaks in the m / z spectrum.

[0104] Based on the mass analysis data processing method described in item 3 and the mass analysis apparatus described in item 8, it is possible to effectively utilize the ion intensity information obtained during measurement to generate a high-precision mass spectrometer.

[0105] (Item 4) The quality analysis data processing method described in any of items 1 through 3 can be set to, further have:

[0106] The condition setting step accepts user input and sets at least one of the envisioned mass range and m / z range for the compound to be analyzed.

[0107] The valence range calculation step uses one or both of the quality range and m / z range set in the condition setting step to calculate the envisioned valence range.

[0108] (Item 9) The quality analysis apparatus described in any one of items 6 to 8 can be configured to further include:

[0109] The condition setting unit accepts user input and sets at least one of the mass range and m / z range envisioned for the compound to be analyzed.

[0110] The price range calculation unit calculates the envisioned price range using one or both of the quality range and the m / z range set by the condition setting unit.

[0111] Expanding the range of hypothetical values ​​increases the reliability of the likelihood of the estimated quality, but correspondingly increases the computational load, computer workload, and processing time. In contrast, the quality analysis data processing method described in item 4 and the quality analysis apparatus described in item 9 allow for limiting the quality range or m / z range based on information known to the user beforehand. This reduces computational costs that contribute little to improving the reliability of the analysis results, thereby lessening the computer load and shortening processing time.

[0112] (Item 5) In any of the quality analysis data processing methods described in items 1 to 4, it can be set such that, in the grade selection step, the likelihood is compared with a predetermined threshold, and if the likelihood exceeds the threshold, it is inferred that the reliability of the estimated quality or grade corresponding to the likelihood is high.

[0113] (Item 10) In any of the quality analysis apparatuses described in items 6 to 9, the grade selection unit can be configured to compare the likelihood with a predetermined threshold, and if the likelihood exceeds the threshold, infer that the reliability of the estimated quality or grade corresponding to the likelihood is high.

[0114] In the mass analysis data processing method described in item 5 and the mass analysis apparatus described in item 10, the likelihood threshold can be predetermined or set by the user. According to the mass analysis data processing method described in item 5 and the mass analysis apparatus described in item 10, an appropriate quality level can be selected, and a high-precision mass spectrum can be generated based on that quality level, or the mass of the target compound can be calculated with high precision.

[0115] Explanation of reference numerals in the attached figures

[0116] 1 Measurement Department

[0117] 10ES I

[0118] 11 Mass Separation Section

[0119] 12 Ion Detection Section

[0120] 2. Analysis and Control Department

[0121] 3 Data Processing Department

[0122] 30MS Analysis Data Collection Department

[0123] 31 Data Storage Department

[0124] 32 Analysis Condition Setting Section

[0125] 33. Price range determination department

[0126] 34 Peak Detection Department

[0127] 35. Preliminary Estimates and Quality Calculation Department

[0128] 36 Likelihood Calculation Department

[0129] 37 Quality Grade Selection Department

[0130] 38 Mass Spectrometry Generation Section

[0131] 4 Input Section

[0132] 5. Display section.

Claims

1. A method for processing quality analysis data, which is a data processing method for processing data obtained through quality analysis of samples, characterized in that, have: The peak information acquisition step involves peak detection on the m / z spectrum based on the acquired data, and collecting peak information containing the m / z values ​​of multiple detected peaks. The preliminary quality calculation step involves multiplying the m / z value of each peak contained in the peak information by multiple values ​​within a predetermined range of assumed values ​​to calculate the preliminary quality. In the grade selection step, for the multiple estimated qualities calculated in the estimated quality calculation step, the frequency of each estimated quality or grade of each specified quality width is calculated as the likelihood of the estimated quality. Based on the likelihood, the estimated quality or grade that is inferred to have high reliability is selected. The inferred quality calculation step involves calculating the inferred quality of a compound corresponding to one or more estimated qualities or grades selected in the grade selection step, based on the peak information and valence of the corresponding peaks when each estimated quality was calculated in the estimated quality calculation step.

2. The quality analysis data processing method as described in claim 1, characterized in that, In addition to the m / z value of the peak, the peak information also includes the intensity value of the peak. The inferred quality calculation step includes a spectrum generation step, which generates a mass spectrum with mass on the horizontal axis and intensity value on the vertical axis based on the peak information and valence of the peaks corresponding to the estimated quality.

3. The quality analysis data processing method as described in claim 2, characterized in that, In the spectrum generation step, a mass spectrum is generated by merging the data with a mass width corresponding to the mass resolution of the peaks in the m / z spectrum.

4. The quality analysis data processing method as described in claim 1, characterized in that, It also has: The condition setting step accepts user input and sets at least one of the envisioned mass range and m / z range for the compound to be analyzed. The valence range calculation step uses one or both of the quality range and m / z range set in the condition setting step to calculate the envisioned valence range.

5. The quality analysis data processing method as described in claim 1, characterized in that, In the grade selection step, the likelihood is compared with a predetermined threshold. If the likelihood exceeds the threshold, it is inferred that the reliability of the estimated quality or grade corresponding to the likelihood is high.

6. A quality analysis device, characterized in that, have: The measurement department acquires data by performing quality analysis on the samples; The peak information acquisition unit performs peak detection on the m / z spectrum based on the data acquired by the measurement unit and collects peak information including the m / z values ​​of multiple detected peaks; The preliminary quality calculation unit calculates the preliminary quality by multiplying the m / z value of each peak contained in the peak information by multiple values ​​within a predetermined range of assumed values. The grade selection unit calculates the frequency of each estimated quality or grade within each specified quality width for multiple estimated qualities calculated by the estimated quality calculation unit as the likelihood of the estimated quality, and selects estimated qualities or grades that are inferred to have high reliability based on the likelihood. The inferred quality calculation unit calculates the inferred quality of the compound corresponding to the estimated quality for one or more estimated qualities or estimated qualities included in the grade selected by the grade selection unit, based on the peak information and valence of the corresponding peak when each estimated quality is calculated by the estimated quality calculation unit.

7. The quality analysis apparatus as described in claim 6, characterized in that, In addition to the m / z value of the peak, the peak information also includes the intensity value of the peak. The inferred quality calculation unit includes a spectrum generation unit, which generates a mass spectrum with mass on the horizontal axis and intensity value on the vertical axis based on the peak information and valence of the peaks corresponding to the estimated quality.

8. The quality analysis apparatus as described in claim 7, characterized in that, The spectrum generation unit generates a mass spectrum by merging data with a mass width corresponding to the mass resolution of the peaks in the m / z spectrum.

9. The quality analysis apparatus as described in claim 6, characterized in that, It also has: The condition setting unit accepts user input and sets at least one of the mass range and m / z range envisioned for the compound to be analyzed. The price range calculation unit calculates the envisioned price range using one or both of the quality range and the m / z range set by the condition setting unit.

10. The quality analysis apparatus as described in claim 6, characterized in that, The grade selection unit compares the likelihood with a predetermined threshold, and if the likelihood exceeds the threshold, it infers that the reliability of the estimated quality or grade corresponding to the likelihood is high.

Images