Metabolite analysis data processing device and metabolite analysis data processing method

Abstract

In the present invention, a multivariate analysis process is performed on data acquired by performing mass spectrometry on samples collected from a plurality of subjects, and the results of the analysis are displayed on a display screen. When an operation for designating one or more specific instruction points is received in a display region for multivariate analysis results on the display screen, a metabolite associated with the one or more instruction points is identified, and a mark indicating that the metabolite is the identified metabolite is inserted into a placement region for the metabolite on a metabolic map displayed on the display screen.

Description

Data processing device for metabolite analysis and data processing method for metabolite analysis

[0001] The present invention relates to a data processing device for metabolite analysis and a method for metabolite analysis.

[0002] Metabolite analysis is performed using chromatography-mass spectrometers such as gas chromatography-mass spectrometers (GC / MS) and liquid chromatography-mass spectrometers (LC / MS) for the diagnosis and treatment of diseases accompanied by metabolic disorders, as well as for the research and development of therapeutic drugs for such diseases. Metabolite analysis using chromatography-mass spectrometers is also performed in pharmacological studies related to the safety and toxicity of pharmaceuticals, health foods, and drugs, as well as in metabolomic analysis of animals and plants.

[0003] In the field of metabolism, charts called metabolic maps are often used to depict metabolic pathways in the body of humans and other organisms. Metabolic maps typically include information on various compounds (metabolites) produced during the metabolic process, chemical reactions, and enzymes involved in metabolism, allowing for a quick understanding of the metabolic process. Such metabolic maps provide a comprehensive understanding of the chain of metabolites produced in the body from ingested food, making it easy to determine the types of metabolites present in bodily fluids such as urine.

[0004] Whether or not a test sample contains various metabolites listed in a metabolic map can be confirmed by analyzing the metabolites contained in the test sample using a chromatograph mass spectrometer. The chromatograph mass spectrometer collects data having three dimensions: retention time, mass-to-charge ratio, and signal intensity. By processing this data, a mass chromatogram showing the relationship between retention time and signal intensity for a specific mass-to-charge ratio, a mass spectrum showing the relationship between mass-to-charge ratio and signal intensity at a specific retention time, and the like are generated. Therefore, the presence of metabolites listed in the metabolic map can be confirmed by identifying the metabolites contained in the test sample from the mass spectrum or mass chromatogram corresponding to the chromatogram peaks obtained by chromatograph mass analysis of the test sample.

[0005] In recent years, in order to confirm the relationship between a specific disease and a metabolite, or the effectiveness of a therapeutic drug for a specific disease, multivariate analysis (multiple types of characteristic values) data related to each other obtained from a large number of samples using a chromatography mass spectrometer has been carried out using techniques such as discriminant analysis, factor analysis, cluster analysis, SIMCA (Soft Independent Modeling of Class Analogy), partial least squares (PLS), O-PLS (Orthogonal PLS), KNN (k nearest neighbors), and principal component analysis (PCA).

[0006] For example, Patent Literature 1 discloses an example in which mass spectrometry data obtained from multiple test samples is subjected to principal component analysis and the results are displayed as a score plot and a loading plot. The score plot obtained as a result of such principal component analysis makes it possible to easily grasp the grouping state of the multiple test samples, and the loading plot makes it possible to grasp how each metabolite contributes to the grouping of the multiple test samples.

[0007] However, the axes of score plots and loading plots are merely axes set for calculation purposes and have no relation to the mass-to-charge ratios or retention times of metabolites, etc. Therefore, even if a user looks at the results of principal component analysis of mass spectrometry data obtained for test samples belonging to multiple groups, it is difficult to obtain information such as the relationship between multiple metabolites that affect the grouping of multiple test samples.

[0008] Japanese Patent Application Laid-Open No. 2009-025056

[0009] The problem that the present invention aims to solve is to make it possible to easily understand the relationships between multiple metabolites that affect the grouping of test samples belonging to multiple groups from the results of multivariate analysis of data obtained by mass spectrometry of the test samples.

[0010] In order to solve the above problems, one embodiment of the present invention provides a data processing device for metabolite analysis, which comprises: an analysis result display processing unit that performs multivariate analysis processing on data obtained by mass spectrometry of samples collected from multiple subjects, and displays the resulting multivariate analysis results on a display screen; a designation operation receiving unit that receives an operation to designate one or more specific indicator points in a display area of ​​the multivariate analysis results on the display screen; a metabolic map data storage unit that stores metabolic map display data for one or more metabolic maps to display a metabolic map on which multiple metabolites are listed; a metabolic map display processing unit that reads out metabolic map display data for a specified metabolic map from the metabolic map data storage unit and displays the specified metabolic map on the display screen; and a metabolite display processing unit that identifies a metabolite associated with the one or more indicator points accepted by the designation operation receiving unit, and inserts a marker indicating the identified metabolite into the display area of ​​the metabolite on the metabolic map displayed on the display screen.

[0011] In addition, a data processing method for metabolite analysis according to another embodiment of the present invention, which has been made to solve the above-mentioned problems, includes an analysis result display step of performing multivariate analysis processing on data acquired by mass spectrometry of samples collected from multiple subjects and displaying the resulting multivariate analysis results on a display screen; a designation operation receiving step of receiving an operation to designate one or more specific indicator points in a display area of ​​the multivariate analysis results on the display screen; a metabolic map display step of reading metabolic map display data for displaying a metabolic map on which multiple metabolites are listed from a metabolic map data storage unit and displaying the specified metabolic map on the display screen; a metabolite identification step of identifying a metabolite associated with the one or more indicator points received in the designation operation receiving step; and a metabolite display processing step of inserting a marker indicating the identified metabolite into a display area of ​​the metabolite corresponding to the identified metabolite on the specified metabolic map displayed on the display screen.

[0012] According to the metabolite analysis data processing device and metabolite analysis data processing method of the present invention, by specifying indicator points that affect the grouping of the multiple test samples from the results of multivariate analysis of data obtained by mass spectrometry of the multiple test samples, metabolites associated with the indicator points that affect the grouping of the multiple test samples are identified, and markers indicating the identified metabolites are displayed on the metabolic map, making it easy to understand the positions of the metabolites on the metabolic map. Therefore, the relationships between the multiple metabolites that affect the grouping of the multiple test samples can be easily understood from the positions on the metabolic map of the multiple metabolites that affect the grouping of the multiple test samples.

[0013] Fig. 1 is a block diagram showing a schematic configuration of a data processing device for metabolite analysis according to the present invention. Fig. 2 shows a score plot (a) and a loading plot (b). Fig. 3 shows a state in which a plurality of indicator points are specified on a loading plot. Fig. 4 shows a state in which a plurality of indicator points are specified on a volcano plot. Fig. 5 shows an example of a metabolic map display before markers indicating metabolites corresponding to the specified indicator points are inserted. Fig. 6 shows an example of a metabolic map display after markers indicating metabolites corresponding to the specified indicator points are inserted.

[0014] An embodiment of a data processing device and method for metabolite analysis according to the present invention will now be described with reference to the accompanying drawings. FIG. 1 is a block diagram showing a schematic configuration of a data processing device for metabolite analysis 100 according to this embodiment. The data processing device for metabolite analysis 100 includes a data analysis unit 1 having, as functional blocks, a metabolic map display data storage unit 10 (corresponding to a metabolic map data storage unit), a metabolite information storage unit 11, a mass analysis data storage unit 12, a peak list creation unit 13, a multivariate analysis unit 14, an analysis result display processing unit 15, a designation operation acceptance unit 16, and a metabolic map display processing unit 17; an input unit 2 through which a user can input or give instructions; and a display unit 3 on which analysis results and a metabolic map are displayed. A predetermined mass spectrometer 4 is connected to the data analysis unit 1, and data acquired therefrom is input thereto.

[0015] Typically, the data analysis unit 1 is a computer such as a personal computer or a more powerful workstation, and the above-mentioned functional blocks are realized by running a dedicated data analysis program installed on the computer. In this case, the input unit 2 is a pointing device such as a computer keyboard or mouse, and the display unit 3 is a PC monitor.

[0016] In this embodiment, the mass spectrometer 4 is a chromatographic mass spectrometer such as a GC / MS or LC / MS, which is connected to the data analysis unit 1. When a predetermined sample is analyzed by this mass spectrometer 4, data (mass analysis data) having three dimensions, namely, retention time, mass-to-charge ratio value, and signal intensity value, is collected. The data collected for one sample is stored as one file in the mass analysis data storage unit 12.

[0017] However, not only data obtained by one specific mass spectrometer 4, but also data similarly obtained by other mass spectrometers can be stored in the mass analysis data storage unit 12. In other words, the mass analysis data for multiple samples to be analyzed in the data analysis unit 1 may be data obtained by one specific mass spectrometer, or may be data obtained by multiple different mass spectrometers, and there are no restrictions on the device that obtains the mass analysis data.

[0018] Mass analysis data for a large number of samples is stored in the mass analysis data storage unit 12. The peak list creation unit 13 detects peaks in mass spectra or mass chromatograms based on individual mass analysis data according to a predetermined algorithm, and determines the positions (mass-to-charge ratio values, retention times) and signal intensities of the detected peaks. Then, for each sample, it creates a peak list that collects a large number of pairs of mass-to-charge ratio values ​​or retention times and signal intensities, and stores the list in the mass analysis data storage unit 12.

[0019] Here, it is assumed that, as a prerequisite for performing multivariate analysis, data files containing peak lists for each of a large number of samples have been created and stored in the mass spectrometry data storage unit 12. For example, if the purpose of performing multivariate analysis is to find metabolic abnormalities observed in patients suffering from a specific disease, the large number of samples will include samples that can be divided into two groups: blood, urine, etc. samples collected from multiple patients, and blood, urine, etc. samples collected from multiple healthy individuals.

[0020] The metabolic map display data storage unit 10 stores metabolic map display data (hereinafter referred to as metabolic map data) for displaying one or more (usually multiple) metabolic maps on the screen of the display unit 3. The metabolic map data is data for displaying the names of multiple metabolites that make up the metabolic map on the screen of the display unit 3, arranged along metabolic pathways.

[0021] The metabolite information storage unit 11 stores, for each metabolite, analysis-related information for metabolites that can be analyzed by the mass spectrometer 4, including analysis conditions including data processing conditions, and analysis results such as the retention index of known standard samples, mass spectra, and characteristic ion information (such as the mass-to-charge ratio of ions and the intensity ratio of multiple ions). The analysis results (mass spectra, retention indices, etc.) of the standard samples included in the analysis-related information for each metabolite may be obtained by actually measuring a standard sample containing the metabolite using the mass spectrometer 4, or information published in existing academic papers may be used. Note that when a certain metabolite is measured using the mass spectrometer 4, it may be measured multiple times under multiple different analysis conditions. Therefore, not only one analysis-related information is associated with a single metabolite, but multiple analysis-related information may also be associated with a single metabolite.

[0022] The multivariate analysis unit 14 generates a covariance matrix, a type of correlation matrix, from data on multiple given samples and calculates loadings corresponding to each compound from the covariance matrix. It also calculates scores for each sample from the data on multiple samples and the loadings calculated as described above. The multivariate analysis unit 14 creates a loadings plot by positioning the calculated loadings on a graph consisting of two axes representing two mutually orthogonal principal components, and creates a score plot by positioning the calculated scores on a graph consisting of two axes representing two mutually orthogonal principal components. Furthermore, the multivariate analysis unit 14 creates a volcano plot that visualizes the ratio of the mean values ​​(fold change) and statistically significant differences (p-values) between the two groups based on peak lists of multiple samples belonging to each of the two given groups.

[0023] Next, we will explain the operations and processing steps for searching for metabolites that characterize a group of multiple samples using the metabolite analysis system configured as described above, as well as the operations and processing steps for displaying the searched metabolites on a metabolic map.

[0024] The user (operator) issues an instruction to display a metabolic map showing a target metabolic pathway by operating the input unit 2. In response to this instruction, the metabolic map display processing unit 17 reads out the corresponding metabolic map data from the metabolic map display data storage unit 10 and displays the metabolic map on the screen of the display unit 3.

[0025] The metabolic map displayed on the screen of the display unit 3 lists the names of metabolites included in the metabolic map along with a rectangular frame 51. As will be described later, a mark 5 indicating a metabolite corresponding to a pointer point designated by the user in the display area of ​​the multivariate analysis results is inserted into the rectangular frame 51, but as shown in Fig. 5, in the initial state, the rectangular frame 51 is blank.

[0026] The user also performs a predetermined operation on the input unit 2 to start the data analysis program, which causes an analysis processing window (not shown) to be displayed on the screen of the display unit 3. Note that either the operation to display the metabolic map on the screen of the display unit 3 or the operation to start the data analysis program can be performed first.

[0027] The analysis processing window has, for example, an analysis result display area and a sample setting area, and the user can set a group of samples to be subjected to analysis processing by performing a predetermined operation in the sample setting area. Any number of groups, two or more, can be set as groups of samples to be subjected to analysis processing. Examples of groups that can be set include two groups, a group of patients and a group of healthy subjects, and a group of men and a group of women.

[0028] Once the sample groups have been set, the multivariate analysis unit 14 reads out the peak list data files corresponding to the samples included in each group from the mass analysis data storage unit 12 and organizes the data in the data files to create a peak matrix. The peak matrix can be, for example, a matrix in which the mass-to-charge ratio values ​​of all peaks are arranged vertically and the samples are arranged horizontally.

[0029] Once the peak matrix has been created, the multivariate analysis unit 14 performs multivariate analysis on this peak matrix. As the multivariate analysis method, any method can be selected from, for example, principal component analysis (PCA), partial least squares (PLS), SIMCA (Soft Independent Modeling of Class Analogy), and orthogonal PLS (O-PLS). The multivariate analysis unit 14 performs multivariate analysis using the selected method and creates a score plot, a loading plot, and a volcano plot as the results of the multivariate analysis.

[0030] The created score plot, loading plot, and volcano plot are all sent to the analysis result display processing unit 15, which displays them on the screen of the display unit 3. Figure 2 shows examples of a score plot (Figure 2(a)) and a loading plot (Figure 2(b)) displayed on the screen of the display unit 3. The score plot and the loading plot may be displayed in the same window on the screen of the display unit 3, or in separate windows.

[0031] Each indicator point on the score plot (shown as a black circle on the score plot) represents a sample. Based on the relative positions of the indicator points, a large number of samples can be divided into multiple groups, as shown in the figure. In the example shown in Figure 2, samples collected from subjects A to G are divided into two groups: "female" and "male."

[0032] Each point on the loadings plot (shown as a black circle on the loadings plot) corresponds to a variable collected by mass spectrometry of the sample, and each point represents a compound (metabolite) contained in at least one sample. On the loadings plot, compounds commonly contained in each sample are clustered near the origin (0,0), while unique compounds that characterize differences between groups are located away from the origin.

[0033] The axes of the loading plot (and score plot as well) are merely calculated axes and have no relation to the retention time or mass-to-charge ratio of each compound. Therefore, even if a user looks at the loading plot, they cannot tell what metabolite each indicator represents. Therefore, the user performs a predetermined operation on the input unit 2 to specify one or more indicators on the loading plot.

[0034] As a method of specifying, when the input unit 2 is, for example, a mouse (or other pointing device), a method of specifying each of the specified points on the loading plot by clicking, or a method of specifying a range including multiple specified points by dragging the mouse, can be used. Figure 3 shows an example of the latter method of specifying, where the range specified by the user is indicated by a rectangular frame with a dashed line and a rectangular frame with a broken line.

[0035] When a certain range (or a specific indicator point) is designated on the loading plot by operating the input unit 2, the designation operation receiving unit 16 reads out mass spectrometry data corresponding to the indicator point for which the designation operation was received from the mass spectrometry data storage unit 12 and inputs the data to the metabolic map display processing unit 17. The metabolic map display processing unit 17 identifies the metabolite indicated by each indicator point based on the input data of each indicator point and the information stored in the metabolite information storage unit 11, and inserts a mark indicating the identified metabolite into a rectangular frame 51 in the region of the metabolic map where the metabolite is displayed. That is, in this embodiment, the metabolic map display processing unit 17 also functions as a metabolite display processing unit. The metabolic map shown in FIG. 6 shows an example in which an elliptical mark 5 is inserted into the rectangular frame 51 as a mark indicating the identified metabolite. However, the form of this mark is arbitrary, and instead of inserting a mark, the display color of the rectangular frame 51 may be changed or the interior of the rectangular frame 51 may be filled in. Alternatively, the name of the metabolite may be enclosed in a rectangular frame, and the display color of the rectangular frame or the interior color of the rectangular frame may be changed as a mark indicating the identified metabolite. In short, it is sufficient if the metabolites corresponding to the indicated points on the loading plot (results of multivariate analysis) can be easily identified on the metabolic map.

[0036] 3 shows an example in which two rectangular frames are set on a loading plot to encompass multiple indicator points representing compounds that are thought to be characteristic of the male and female groups, respectively. Once the rectangular frames are set in this manner, indicator points within the rectangular frames are designated, and markers representing metabolites identified from the designated indicator points are inserted into the metabolic map. This allows the user to infer that the metabolites that characterize the differences between the male and female groups of subjects are involved in a certain metabolic pathway on the metabolic map.

[0037] As described above, the metabolite analysis data processing device 100 of this embodiment can easily grasp the positions on a metabolic map of metabolites that characterize differences between groups identified from the results of multivariate analysis (principal component analysis) of mass spectrometry data of samples from multiple groups. Therefore, the metabolite analysis data processing device 100 of this embodiment can be useful, for example, for searching for markers that lead to the early detection of patients suffering from a specific disease.

[0038] While the above describes an example of specifying one or more indicator points on a loading plot, one or more indicator points can also be specified on a volcano plot such as that shown in FIG. 4. Each indicator point on a volcano plot represents a metabolite contained in at least one sample, with the higher the point on the vertical axis, the more significantly different the metabolite is between the two groups, and the further away from the center on the horizontal axis, the greater the difference in average values ​​between the two groups. Therefore, when a volcano plot is created as a result of multivariate analysis, the user can easily specify indicator points (metabolites) that affect the differences between the groups by viewing the volcano plot displayed on the screen of the display unit 3.

[0039] Aspects It will be apparent to those skilled in the art that the exemplary embodiments described above are examples of the following aspects.

[0040] (Item 1) A data processing device for metabolite analysis according to one aspect of the present invention comprises: an analysis result display processing unit that performs multivariate analysis processing on data obtained by mass spectrometry of samples collected from multiple subjects and displays the resulting multivariate analysis results on a display screen; a designation operation receiving unit that receives an operation to designate one or more specific indicator points in a display area of ​​the multivariate analysis results on the display screen; a metabolic map data storage unit that stores metabolic map display data for one or more metabolic maps to display a metabolic map on which multiple metabolites are listed; a metabolic map display processing unit that reads out metabolic map display data for a specified metabolic map from the metabolic map data storage unit and displays the specified metabolic map on the display screen; and a metabolite display processing unit that identifies a metabolite associated with the one or more indicator points accepted by the designation operation receiving unit and inserts a marker indicating the identified metabolite into the display area of ​​the metabolite on the metabolic map displayed on the display screen.

[0041] In the metabolite analysis data processing device according to paragraph 1, an analysis result display processor performs multivariate analysis on data obtained by mass spectrometry of samples collected from multiple subjects, such as mass spectrum data, mass chromatogram data, or peak data obtained by performing peak detection on the mass spectrum or mass chromatogram according to a predetermined algorithm, and displays the results on a display screen. Multivariate analysis is a statistical technique for analyzing the interrelationships between multiple variables based on data related to the variables. The results of the multivariate analysis displayed on the display screen can be used to group multiple samples or to search for metabolites that contribute to the grouping of the multiple samples. In other words, the user's operation of specifying one or more specific indicator points in the display area of ​​the multivariate analysis results on the display screen is equivalent to selecting a specific grouped sample or a metabolite that contributes to the grouping of the multiple samples. When the user's operation to specify an indication point is accepted by the specification operation accepting unit, the metabolite associated with that indication point is identified, and a mark indicating that it is the specified metabolite is inserted at the location of the specified metabolite among the metabolites listed on the metabolic map displayed on the display screen, making it easy to grasp the position of the metabolite on the metabolic map.

[0042] (2) The metabolite analysis data processing device according to 2 may be the metabolite analysis data processing device according to 1, wherein the mass spectrometer is a gas chromatograph mass spectrometer or a liquid chromatograph mass spectrometer.

[0043] (Item 3) The data processing device for metabolite analysis according to item 3 is the data processing device for metabolite analysis according to item 1 or 2, wherein the multivariate analysis process can be any one of Principal Component Analysis (PCA), Partial Least Squares (PLS), Soft Independent Modeling of Class Analogy (SIMCA), and Orthogonal PLS (O-PLS).

[0044] (4) The data processing device for metabolite analysis according to 4 is the data processing device for metabolite analysis according to any one of 1 to 3, wherein the multivariate analysis results displayed on the display screen by the analysis result display processing unit can be a loading plot or a volcano plot.

[0045] (Item 5) A data processing method for metabolite analysis according to one aspect of the present invention includes an analysis result display step of performing multivariate analysis processing on data obtained by mass spectrometry of samples collected from multiple subjects and displaying the resulting multivariate analysis results on a display screen; a designation operation receiving step of receiving an operation to designate one or more specific indicator points in a display area of ​​the multivariate analysis results on the display screen; a metabolic map display step of reading metabolic map display data for displaying a metabolic map on which multiple metabolites are listed from a metabolic map data storage unit and displaying the specified metabolic map on the display screen; a metabolite identification step of identifying a metabolite associated with the one or more indicator points received in the designation operation receiving step; and a metabolite display processing step of inserting a marker indicating the identified metabolite into a display area of ​​the metabolite corresponding to the identified metabolite on the specified metabolic map displayed on the display screen.

[0046] DESCRIPTION OF SYMBOLS 1... Data analysis section 10... Metabolic map display data storage section 11... Metabolite information storage section 12... Mass analysis data storage section 13... Peak list creation section 14... Multivariate analysis section 15... Analysis result display processing section 16... Designation operation reception section 17... Metabolic map display processing section 2... Input section 3... Display section 4... Mass spectrometer 5... Mark 100... Metabolic analysis data processing device

Claims

1. A data processing device for metabolite analysis, comprising: an analysis result display processing unit that performs multivariate analysis processing on data obtained by mass spectrometry of samples collected from multiple subjects and displays the resulting multivariate analysis results on a display screen; a designation operation receiving unit that receives an operation to designate one or more specific indicator points in the display area of ​​the multivariate analysis results on the display screen; a metabolic map data storage unit that stores metabolic map display data for one or more metabolic maps to display a metabolic map on which multiple metabolites are listed; a metabolic map display processing unit that reads out metabolic map display data for a specified metabolic map from the metabolic map data storage unit and displays the specified metabolic map on the display screen; and a metabolite display processing unit that identifies a metabolite associated with the one or more indicator points accepted by the designation operation receiving unit and inserts a marker indicating the identified metabolite into the area on the metabolic map displayed on the display screen where the metabolite is listed.

2. A data processing device for metabolite analysis according to claim 1, wherein the mass spectrometer is a gas chromatograph mass spectrometer or a liquid chromatograph mass spectrometer.

3. A data processing device for metabolite analysis according to claim 1, wherein the multivariate analysis process is any one of principal component analysis (PCA), partial least squares (PLS), soft independent modeling of class analogy (SIMCA), and orthogonal PLS (O-PLS).

4. A data processing device for metabolite analysis according to claim 1, wherein the multivariate analysis results displayed on the display screen by the analysis result display processing section are a loading plot or a volcano plot.

5. A data processing method for metabolite analysis, comprising: an analysis result display step for performing multivariate analysis processing on data obtained by mass spectrometry of samples collected from a plurality of subjects, and displaying the resulting multivariate analysis results on a display screen; a designation operation receiving step for receiving an operation to designate one or more specific indicator points in a display area of ​​the multivariate analysis results on the display screen; a metabolic map display step for reading metabolic map display data for displaying a metabolic map containing a plurality of metabolites from a metabolic map data storage unit, and displaying the specified metabolic map on the display screen; a metabolite identification step for identifying a metabolite associated with the one or more indicator points received in the designation operation receiving step; and a metabolite display processing step for inserting a marker indicating the identified metabolite into a display area of ​​the metabolite corresponding to the identified metabolite on the specified metabolic map displayed on the display screen.

Images