A multidimensional validation method for metabolomics variable analysis data

Abstract

The present invention relates to the field of metabolomics variable analysis, and in particular to a multidimensional verification method for metabolomics variable analysis data, comprising: S1, preprocessing metabolomics variable analysis data to obtain metabolomics variable analysis preprocessing data; S2, establishing multidimensional verification marker characteristics of the metabolomics variable analysis data using the metabolomics variable analysis preprocessing data; S3, performing linear regression verification based on the multidimensional verification marker characteristics using the metabolomics variable analysis preprocessing data to obtain a multidimensional verification result, through data preprocessing, establishing a verification starting point for the metabolomics data based on different properties of single variables and multivariates, and reasonably relying on a metabolomics model to generate a screening marker, which is both associated with the analysis data to be verified and avoids inaccurate verification results caused by excessive association.

Description

Technical Field

[0001] The present invention relates to the field of metabolomics variable analysis, and in particular to a multidimensional verification method for metabolomics variable analysis data. Background Art

[0002] Metabolomics is a newly developed discipline in the field of systems biology following genomics and proteomics. It explores the metabolic mechanism of the entire organism by detecting the changes in metabolites in the organism after exogenous stimulation or genetic modification. Its research object is all endogenous small molecule metabolites (molecular weight <1000Da) in the organism.

[0003] The research methods are high-throughput detection technology and data processing methods, and the ultimate goal is data modeling and biomarker screening. Metabolomics research generates a large amount of data. These data have complex characteristics such as high dimension, small sample, and high noise. Specifically, the physical and chemical properties of different metabolites vary greatly, and the dynamic range of their concentration content is as wide as 7 to 9 orders of magnitude. At the same time, there are various sources of variation between biological individuals. For example, age and gender may affect the changes in metabolites. Third, instrument measurement is affected by various factors and is prone to random measurement errors and systematic errors. This makes it extremely difficult to identify important biomarkers. In addition, various metabolites may not only have simple additive effects, but also may have interactions, which increases the difficulty of identifying these biomarkers with complex relationships. Therefore, a practical and widely applicable verification method is urgently needed. Summary of the invention

[0004] In view of the shortcomings of the prior art, the present invention provides a multidimensional verification method for metabolomics variable analysis data, which can quickly and effectively realize the output verification of metabolomics analysis data by separating and verifying univariate data and multivariate data.

[0005] To achieve the above object, the present invention provides a multidimensional verification method for metabolomics variable analysis data, comprising:

[0006] S1. Preprocessing the metabolomics variable analysis data to obtain metabolomics variable analysis preprocessing data;

[0007] S2. Using the metabolomics variable analysis preprocessing data to establish a multidimensional verification marker feature of the metabolomics variable analysis data;

[0008] S3. Utilizing the metabolomics variable analysis preprocessed data, linear regression verification is performed according to the multidimensional verification marker characteristics to obtain multidimensional verification results.

[0009] In this embodiment, a multidimensional verification method for metabolomics variable analysis data is provided. Metabolomics is an emerging discipline that analyzes all low molecular weight metabolites of a certain organism or cell (with organic and inorganic metabolites with a relative molecular mass <1000 as the core research area). After biological samples are analyzed and detected by high-throughput instruments such as NMR, GC-MS, and LC-MS, a large amount of data can be generated. These data have complex characteristics such as high dimension, small sample size, and high noise. At the same time, there are many metabolites and the metabolites are closely related.

[0010] Preferably, the step of preprocessing the metabolomics variable analysis data to obtain metabolomics variable analysis preprocessed data comprises:

[0011] S1-1, determining whether the metabolomics variable analysis data is univariate analysis data, if so, performing correlation analysis using the metabolomics variable analysis data to obtain metabolomics variable analysis preprocessing data, otherwise, executing S1-2;

[0012] S1-2, determining whether the metabolomics variable analysis data is multivariate analysis data, if so, executing S1-3, otherwise, abandoning the processing;

[0013] S1-3, determining whether the metabolomics variable analysis data corresponds to the same metabolomics starting material data, if so, dividing the metabolomics variable analysis data according to the corresponding data dimension to obtain metabolomics variable analysis preprocessing data, otherwise, abandoning the processing.

[0014] In this embodiment, a multidimensional validation method for metabolomics variable analysis data is provided, wherein the metabolomics starting material data corresponds to the starting primer of the current metabolomics model.

[0015] Furthermore, using the metabolomics variable analysis data to perform correlation analysis to obtain metabolomics variable analysis preprocessing data includes:

[0016] S1-1-1, obtaining the starting material data of the metabolomics variable analysis data as the starting point of the correlation analysis;

[0017] S1-1-2, obtaining the model nodes corresponding to the metabolomics analysis data according to the association analysis starting point to establish a model structure node set;

[0018] S1-1-3, determine whether the adjacent subsets of the model structure node set correspond, if so, execute S1-1-4, otherwise, return to S1-1-2;

[0019] S1-1-4, determining whether there is a unique subset of the model structure node set corresponding to the starting point of the correlation analysis, if so, obtaining the corresponding data of the model structure node according to the model structure node set as the metabolomics variable analysis preprocessing data, otherwise, returning to S1-1-1;

[0020] Wherein, each adjacent subset of the model structure node set corresponds to each adjacent subset of the model structure node set and the flow direction of the metabolomics data.

[0021] Preferably, using the metabolomics variable analysis preprocessing data to establish a multidimensional verification marker feature of the metabolomics variable analysis data includes:

[0022] S2-1, obtaining the starting material data corresponding to the preprocessed data of metabolomics variable analysis as the starting features;

[0023] S2-2, determining whether the metabolomics variable analysis preprocessing data is univariate data, if so, obtaining a multidimensional verification marker feature of the metabolomics variable analysis data using a corresponding model of the metabolomics variable analysis data according to the initial feature, otherwise, executing S2-3;

[0024] S2-3, obtaining the approximate metabolite data of the preprocessed data for metabolomics variable analysis;

[0025] S2-4. Obtain multidimensional verification marker features of metabolomics variable analysis data based on the approximate situation of the metabolite data.

[0026] Furthermore, the multidimensional verification marker features of the metabolomics variable analysis data obtained by using the metabolomics variable analysis data corresponding model according to the initial features include:

[0027] S2-2-1, obtaining the metabolite change time of the model corresponding to the metabolomics variable analysis data;

[0028] S2-2-2, determining whether there is a metabolite hysteresis state at the metabolite change time, if so, executing S2-2-3, otherwise, using the metabolite change time as a multidimensional verification marker feature of metabolomics variable analysis data;

[0029] S2-2-3, using the metabolite change time in the metabolite hysteresis state to divide the twin change time and the remaining metabolite change time as multidimensional verification marker features of metabolomics variable analysis data;

[0030] The metabolite change time is the time when the metabolite in the model changes, and the metabolite hysteresis state is the presence of the same metabolite at the same model node at different times in the model corresponding to the metabolomics variable analysis data.

[0031] Furthermore, the approximate situation of metabolite data of obtaining metabolomics variable analysis preprocessing data includes:

[0032] S2-3-1, determining whether the number of starting material data corresponding to the metabolomics variable analysis preprocessing data is greater than 30% of the number of metabolite data, if so, executing S2-3-2, otherwise, abandoning the processing;

[0033] S2-3-2, judging whether the metabolite data of the metabolomics variable analysis preprocessing data have the same metabolite data, if so, executing S2-3-3, otherwise, the metabolite data of the metabolomics variable analysis preprocessing data are similar to non-existent;

[0034] S2-3-3, determine whether the metabolite data corresponds to the same metabolomics variable analysis preprocessing data, if so, merge the same metabolomics variable analysis preprocessing data with the corresponding metabolite data, and return to S2-3-1, otherwise, use the metabolomics variable analysis preprocessing data to establish a similar product time mapping with the corresponding time of the metabolite data as the metabolite data approximation of the metabolite variable analysis preprocessing data.

[0035] Furthermore, obtaining multidimensional verification marker features of metabolomics variable analysis data based on the approximate situation of the metabolite data includes:

[0036] S2-4-1, determining whether the metabolite data approximation exists, if so, using the metabolite data approximation to establish a metabolite time series set of metabolomics variable analysis data as a multi-dimensional verification marker feature, otherwise, executing S2-4-2;

[0037] S2-4-2. Determine whether the metabolomics variable analysis data has the same final metabolite. If so, return to S2-3-1. Otherwise, use the final metabolite as a multi-dimensional verification marker feature.

[0038] Furthermore, the metabolomics variable analysis preprocessed data is used to perform linear regression verification according to the multidimensional verification marker characteristics to obtain multidimensional verification results including:

[0039] S3-1, when the metabolomics variable analysis preprocessing data corresponds to univariate data, using the multidimensional verification marker feature to perform time series linear verification on the metabolomics variable analysis preprocessing data to obtain a multidimensional verification result;

[0040] S3-2. When the metabolomics variable analysis preprocessing data corresponds to multivariate data, the multidimensional verification marker feature is used to perform output regression verification on the metabolomics variable analysis preprocessing data to obtain a multidimensional verification result.

[0041] Furthermore, the multidimensional validation marker features are used to perform time series linear validation on the metabolomics variable analysis preprocessing data to obtain multidimensional validation results including:

[0042] S3-1-1, judging whether the multi-dimensional verification marker characteristics are completely consistent with the corresponding model of the metabolomics variable analysis data, if so, the multi-dimensional verification result is successful, otherwise, directly executing S3-1-2;

[0043] S3-1-2, determine whether the multi-dimensional verification mark feature has a twin change moment, if so, execute S3-1-3, otherwise, the multi-dimensional verification result fails;

[0044] S3-1-3. Determine whether the time series corresponding to the multidimensional verification marker feature is consistent with the time series corresponding to the metabolite of the metabolomics variable analysis preprocessing data. If so, the multidimensional verification result is successful, otherwise, the multidimensional verification result fails.

[0045] Furthermore, the multidimensional validation marker features are used to perform output regression validation on the metabolomics variable analysis preprocessing data to obtain multidimensional validation results including:

[0046] S3-2-1. Determine whether the multi-dimensional verification marker feature has metabolite data similarity. If so, execute S3-2-2. Otherwise, directly execute S3-2-3.

[0047] S3-2-2, judging whether the approximate situation of the metabolite data conforms to the corresponding model of the metabolomics variable analysis preprocessing data, if so, the multidimensional verification result is successful, otherwise, executing S3-2-3;

[0048] S3-2-3, determine whether the multi-dimensional verification marker feature is the final metabolite, if so, execute S3-2-4, otherwise, execute S3-2-5;

[0049] S3-2-4, judging whether the multidimensional validation marker characteristics correspond to the data dimensions of the metabolomics variable analysis preprocessing data, if so, the multidimensional validation result is successful, otherwise, the multidimensional validation result fails;

[0050] S3-2-5. Determine whether the multidimensional verification marker feature corresponds to the metabolite product time of the corresponding model of the metabolomics variable analysis preprocessing data. If so, the multidimensional verification result is successful, otherwise, the multidimensional verification result fails.

[0051] Compared with the closest prior art, the present invention has the following beneficial effects:

[0052] Through data preprocessing, the metabolomics data established a verification starting point based on the different properties of single variables and multivariate data, and reasonably relied on the metabolomics model to generate screening markers, which were both associated with the analytical data to be verified and avoided inaccurate verification results caused by excessive correlation. At the same time, an innovative time series verification marker based on the product moment of the node in the model was proposed, revealing the complex interaction relationship between variables, and finally summarized into two types of inspection data, so as to obtain the final analysis data verification results simply, quickly and effectively. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 The present invention provides a flow chart of a multidimensional verification method for metabolomics variable analysis data. DETAILED DESCRIPTION

[0054] The specific implementation modes of the present invention will be further described in detail below in conjunction with the accompanying drawings.

[0055] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. Example

[0056] The present invention provides a multidimensional verification method for metabolomics variable analysis data, such as Figure 1 As shown, including:

[0057] S1. Preprocessing the metabolomics variable analysis data to obtain metabolomics variable analysis preprocessing data;

[0058] S2. Using the metabolomics variable analysis preprocessing data to establish a multidimensional verification marker feature of the metabolomics variable analysis data;

[0059] S3. Utilizing the metabolomics variable analysis preprocessed data, linear regression verification is performed according to the multidimensional verification marker characteristics to obtain multidimensional verification results.

[0060] S1 specifically includes:

[0061] S1-1, determining whether the metabolomics variable analysis data is univariate analysis data, if so, performing correlation analysis using the metabolomics variable analysis data to obtain metabolomics variable analysis preprocessing data, otherwise, executing S1-2;

[0062] S1-2, determining whether the metabolomics variable analysis data is multivariate analysis data, if so, executing S1-3, otherwise, abandoning the processing;

[0063] S1-3, determining whether the metabolomics variable analysis data corresponds to the same metabolomics starting material data, if so, dividing the metabolomics variable analysis data according to the corresponding data dimension to obtain metabolomics variable analysis preprocessing data, otherwise, abandoning the processing.

[0064] In this embodiment, a multidimensional verification method for metabolomics variable analysis data is provided. The analysis data is classified according to single variables or multivariates. This is because metabolomics data is generally difficult to meet the conditions for parameter testing. At the same time, considering the high-dimensional data generated by metabolomics, different ideas are implemented in the subsequent processing of single variable or multivariate data.

[0065] S1-1 specifically includes:

[0066] S1-1-1, obtaining the starting material data of the metabolomics variable analysis data as the starting point of the correlation analysis;

[0067] S1-1-2, obtaining the model nodes corresponding to the metabolomics analysis data according to the association analysis starting point to establish a model structure node set;

[0068] S1-1-3, determine whether the adjacent subsets of the model structure node set correspond, if so, execute S1-1-4, otherwise, return to S1-1-2;

[0069] S1-1-4, determining whether there is a unique subset of the model structure node set corresponding to the starting point of the correlation analysis, if so, obtaining the corresponding data of the model structure node according to the model structure node set as the metabolomics variable analysis preprocessing data, otherwise, returning to S1-1-1;

[0070] Wherein, each adjacent subset of the model structure node set corresponds to each adjacent subset of the model structure node set and the flow direction of the metabolomics data.

[0071] In this embodiment, a multidimensional verification method for metabolomics variable analysis data is provided. The existence of a unique subset of the model structure node set corresponding to the starting point of the correlation analysis can be understood as that the univariate metabolomics data has a single flow in the model and the node connection is unique. Based on this feature, the current step is quickly verified and screened.

[0072] S2 specifically includes:

[0073] S2-1, obtaining the starting material data corresponding to the preprocessed data of metabolomics variable analysis as the starting features;

[0074] S2-2, determining whether the metabolomics variable analysis preprocessing data is univariate data, if so, obtaining a multidimensional verification marker feature of the metabolomics variable analysis data using a corresponding model of the metabolomics variable analysis data according to the initial feature, otherwise, executing S2-3;

[0075] S2-3, obtaining the approximate metabolite data of the preprocessed data for metabolomics variable analysis;

[0076] S2-4. Obtain multidimensional verification marker features of metabolomics variable analysis data based on the approximate situation of the metabolite data.

[0077] S2-2 specifically includes:

[0078] S2-2-1, obtaining the metabolite change time of the model corresponding to the metabolomics variable analysis data;

[0079] S2-2-2, determining whether there is a metabolite hysteresis state at the metabolite change time, if so, executing S2-2-3, otherwise, using the metabolite change time as a multidimensional verification marker feature of metabolomics variable analysis data;

[0080] S2-2-3, using the metabolite change time in the metabolite hysteresis state to divide the twin change time and the remaining metabolite change time as multidimensional verification marker features of metabolomics variable analysis data;

[0081] The metabolite change time is the time when the metabolite in the model changes, and the metabolite hysteresis state is the presence of the same metabolite at the same model node at different times in the model corresponding to the metabolomics variable analysis data.

[0082] S2-3 specifically includes:

[0083] S2-3-1, determining whether the number of starting material data corresponding to the metabolomics variable analysis preprocessing data is greater than 30% of the number of metabolite data, if so, executing S2-3-2, otherwise, abandoning the processing;

[0084] S2-3-2, judging whether the metabolite data of the metabolomics variable analysis preprocessing data have the same metabolite data, if so, executing S2-3-3, otherwise, the metabolite data of the metabolomics variable analysis preprocessing data are similar to non-existent;

[0085] S2-3-3, determine whether the metabolite data corresponds to the same metabolomics variable analysis preprocessing data, if so, merge the same metabolomics variable analysis preprocessing data with the corresponding metabolite data, and return to S2-3-1, otherwise, use the metabolomics variable analysis preprocessing data to establish a similar product time mapping with the corresponding time of the metabolite data as the metabolite data approximation of the metabolite variable analysis preprocessing data.

[0086] In this embodiment, a multidimensional verification method for metabolomics variable analysis data is provided. When the number of starting material data corresponding to the metabolomics variable analysis preprocessing data is greater than 30% of the number of metabolite data, it can be regarded as a high-dimensional, small sample situation. The number of metabolites is much greater than the number of samples, and it is not suitable to be analyzed using traditional statistical methods. Multivariate analysis is prone to overfitting and dimensionality disaster problems. Therefore, the processing is abandoned in actual operation and the model is adjusted.

[0087] S2-4 specifically includes:

[0088] S2-4-1, determining whether the metabolite data approximation exists, if so, using the metabolite data approximation to establish a metabolite time series set of metabolomics variable analysis data as a multi-dimensional verification marker feature, otherwise, executing S2-4-2;

[0089] S2-4-2. Determine whether the metabolomics variable analysis data has the same final metabolite. If so, return to S2-3-1. Otherwise, use the final metabolite as a multi-dimensional verification marker feature.

[0090] S3 specifically includes:

[0091] S3-1, when the metabolomics variable analysis preprocessing data corresponds to univariate data, using the multidimensional verification marker feature to perform time series linear verification on the metabolomics variable analysis preprocessing data to obtain a multidimensional verification result;

[0092] S3-2. When the metabolomics variable analysis preprocessing data corresponds to multivariate data, the multidimensional verification marker feature is used to perform output regression verification on the metabolomics variable analysis preprocessing data to obtain a multidimensional verification result.

[0093] In this embodiment, a multidimensional verification method for metabolomics variable analysis data is provided. In this verification step, features and two variable data are used for judgment, so that successful analysis data can be quickly screened.

[0094] S3-1 specifically includes:

[0095] S3-1-1, judging whether the multi-dimensional verification marker characteristics are completely consistent with the corresponding model of the metabolomics variable analysis data, if so, the multi-dimensional verification result is successful, otherwise, directly executing S3-1-2;

[0096] S3-1-2, determine whether the multi-dimensional verification mark feature has a twin change moment, if so, execute S3-1-3, otherwise, the multi-dimensional verification result fails;

[0097] S3-1-3. Determine whether the time series corresponding to the multidimensional verification marker feature is consistent with the time series corresponding to the metabolite of the metabolomics variable analysis preprocessing data. If so, the multidimensional verification result is successful, otherwise, the multidimensional verification result fails.

[0098] In this embodiment, a multidimensional verification method for metabolomics variable analysis data is provided. In a single variable data type, the model corresponds to the time of change of each contained node, and the change multiple value of the metabolite concentration at different times can be derived, thereby obtaining a judgment result.

[0099] S3-2 specifically includes:

[0100] S3-2-1. Determine whether the multi-dimensional verification marker feature has metabolite data similarity. If so, execute S3-2-2. Otherwise, directly execute S3-2-3.

[0101] S3-2-2, judging whether the approximate situation of the metabolite data conforms to the corresponding model of the metabolomics variable analysis preprocessing data, if so, the multidimensional verification result is successful, otherwise, executing S3-2-3;

[0102] S3-2-3, determine whether the multi-dimensional verification marker feature is the final metabolite, if so, execute S3-2-4, otherwise, execute S3-2-5;

[0103] S3-2-4, judging whether the multidimensional validation marker characteristics correspond to the data dimensions of the metabolomics variable analysis preprocessing data, if so, the multidimensional validation result is successful, otherwise, the multidimensional validation result fails;

[0104] S3-2-5. Determine whether the multidimensional verification marker feature corresponds to the metabolite product time of the corresponding model of the metabolomics variable analysis preprocessing data. If so, the multidimensional verification result is successful, otherwise, the multidimensional verification result fails.

[0105] In this embodiment, a multi-dimensional verification method for metabolomics variable analysis data is provided. Under the premise that the starting materials are certain, the types of metabolites will not exceed their corresponding categories. If they exceed the categories, the results can be determined to be abnormal.

[0106] In this embodiment, a multidimensional validation method for metabolomics variable analysis data is provided, wherein the data dimension is defined as one data expressing one meaning, and a group of data expressing one or more meanings. Dimension: the organizational form of a group of data (one-dimensional, two-dimensional or multi-dimensional).

[0107] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0108] The present invention is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0109] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0110] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0111] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A multidimensional validation method for metabolomics variable analysis data, characterized in that: include: S1. Preprocessing the metabolomics variable analysis data to obtain metabolomics variable analysis preprocessing data; S2. Using the metabolomics variable analysis preprocessing data to establish a multidimensional verification marker feature of the metabolomics variable analysis data; S3, using the metabolomics variable analysis preprocessed data to perform linear regression verification according to the multidimensional verification marker characteristics to obtain a multidimensional verification result; The method of preprocessing the metabolomics variable analysis data to obtain the metabolomics variable analysis preprocessed data comprises: S1-1, determining whether the metabolomics variable analysis data is univariate analysis data, if so, performing correlation analysis using the metabolomics variable analysis data to obtain metabolomics variable analysis preprocessing data, otherwise, executing S1-2; S1-2, determining whether the metabolomics variable analysis data is multivariate analysis data, if so, executing S1-3, otherwise, abandoning the processing; S1-3, determining whether the metabolomics variable analysis data corresponds to the same metabolomics starting material data, if so, dividing the metabolomics variable analysis data according to the corresponding data dimension to obtain metabolomics variable analysis preprocessing data, otherwise, abandoning the processing; The multidimensional verification marker features of metabolomics variable analysis data established by using the metabolomics variable analysis preprocessing data include: S2-1, obtaining the starting material data corresponding to the preprocessed data of metabolomics variable analysis as the starting features; S2-2, determining whether the metabolomics variable analysis preprocessing data is univariate data, if so, obtaining a multidimensional verification marker feature of the metabolomics variable analysis data using a corresponding model of the metabolomics variable analysis data according to the initial feature, otherwise, executing S2-3; S2-3, obtaining the approximate metabolite data of the preprocessed data for metabolomics variable analysis; S2-4, obtaining multidimensional verification marker features of metabolomics variable analysis data according to the approximate situation of the metabolite data; The multidimensional verification marker features of the metabolomics variable analysis data obtained by using the metabolomics variable analysis data corresponding model according to the initial features include: S2-2-1, obtaining the metabolite change time of the model corresponding to the metabolomics variable analysis data; S2-2-2, determining whether there is a metabolite hysteresis state at the metabolite change time, if so, executing S2-2-3, otherwise, using the metabolite change time as a multidimensional verification marker feature of metabolomics variable analysis data; S2-2-3, using the metabolite change time in the metabolite hysteresis state to divide the twin change time and the remaining metabolite change time as multidimensional verification marker features of metabolomics variable analysis data; The metabolite change time is the time when the metabolite in the model changes, and the metabolite hysteresis state is the presence of the same metabolite at the same model node at different times in the model corresponding to the metabolomics variable analysis data; The metabolomics variable analysis preprocessed data was used to perform linear regression verification based on the multidimensional verification marker characteristics to obtain multidimensional verification results including: S3-1, when the metabolomics variable analysis preprocessing data corresponds to univariate data, using the multidimensional verification marker feature to perform time series linear verification on the metabolomics variable analysis preprocessing data to obtain a multidimensional verification result; S3-2. When the metabolomics variable analysis preprocessing data corresponds to multivariate data, the multidimensional verification marker feature is used to perform output regression verification on the metabolomics variable analysis preprocessing data to obtain a multidimensional verification result.

2. A multidimensional validation method for metabolomics variable analysis data according to claim 1, characterized in that: The metabolomics variable analysis data is used to perform correlation analysis to obtain metabolomics variable analysis preprocessing data, including: S1-1-1, obtaining the starting material data of the metabolomics variable analysis data as the starting point of the correlation analysis; S1-1-2, obtaining the model nodes corresponding to the metabolomics analysis data according to the association analysis starting point to establish a model structure node set; S1-1-3, determine whether the adjacent subsets of the model structure node set correspond, if so, execute S1-1-4, otherwise, return to S1-1-2; S1-1-4, determining whether there is a unique subset of the model structure node set corresponding to the starting point of the correlation analysis, if so, obtaining the corresponding data of the model structure node according to the model structure node set as the metabolomics variable analysis preprocessing data, otherwise, returning to S1-1-1; Wherein, each adjacent subset of the model structure node set corresponds to each adjacent subset of the model structure node set and the flow direction of the metabolomics data.

3. A multidimensional validation method for metabolomics variable analysis data according to claim 1, characterized in that: The approximate situation of obtaining metabolite data of metabolomics variable analysis preprocessing data includes: S2-3-1, determining whether the number of starting material data corresponding to the metabolomics variable analysis preprocessing data is greater than 30% of the number of metabolite data, if so, executing S2-3-2, otherwise, abandoning the processing; S2-3-2, judging whether the metabolite data of the metabolomics variable analysis preprocessing data have the same metabolite data, if so, executing S2-3-3, otherwise, the metabolite data of the metabolomics variable analysis preprocessing data are similar to non-existent; S2-3-3, determine whether the metabolite data corresponds to the same metabolomics variable analysis preprocessing data, if so, merge the same metabolomics variable analysis preprocessing data with the corresponding metabolite data, and return to S2-3-1, otherwise, use the metabolomics variable analysis preprocessing data to establish a similar product time mapping with the corresponding time of the metabolite data as the metabolite data approximation of the metabolite variable analysis preprocessing data.

4. A multidimensional validation method for metabolomics variable analysis data according to claim 3, characterized in that: The multi-dimensional validation marker features of metabolomics variable analysis data obtained based on the approximate situation of the metabolite data include: S2-4-1, determining whether the metabolite data approximation exists, if so, using the metabolite data approximation to establish a metabolite time series set of metabolomics variable analysis data as a multi-dimensional verification marker feature, otherwise, executing S2-4-2; S2-4-2. Determine whether the metabolomics variable analysis data has the same final metabolite. If so, return to S2-3-1. Otherwise, use the final metabolite as a multi-dimensional verification marker feature.

5. A multidimensional validation method for metabolomics variable analysis data according to claim 1, characterized in that: The multidimensional validation marker features are used to perform time series linear validation on the metabolomics variable analysis preprocessing data to obtain multidimensional validation results including: S3-1-1, judging whether the multi-dimensional verification marker characteristics are completely consistent with the corresponding model of the metabolomics variable analysis data, if so, the multi-dimensional verification result is successful, otherwise, directly executing S3-1-2; S3-1-2, determine whether the multi-dimensional verification mark feature has a twin change moment, if so, execute S3-1-3, otherwise, the multi-dimensional verification result fails; S3-1-3. Determine whether the time series corresponding to the multidimensional verification marker feature is consistent with the time series corresponding to the metabolite of the metabolomics variable analysis preprocessing data. If so, the multidimensional verification result is successful, otherwise, the multidimensional verification result fails.

6. A multidimensional validation method for metabolomics variable analysis data according to claim 5, characterized in that: The multidimensional validation marker features are used to perform output regression validation on the metabolomics variable analysis preprocessing data to obtain multidimensional validation results including: S3-2-1. Determine whether the multi-dimensional verification marker feature has metabolite data similarity. If so, execute S3-2-2. Otherwise, directly execute S3-2-3. S3-2-2, judging whether the approximate situation of the metabolite data conforms to the corresponding model of the metabolomics variable analysis preprocessing data, if so, the multidimensional verification result is successful, otherwise, executing S3-2-3; S3-2-3, determine whether the multi-dimensional verification marker feature is the final metabolite, if so, execute S3-2-4, otherwise, execute S3-2-5; S3-2-4, judging whether the multidimensional validation marker characteristics correspond to the data dimensions of the metabolomics variable analysis preprocessing data, if so, the multidimensional validation result is successful, otherwise, the multidimensional validation result fails; S3-2-5. Determine whether the multidimensional verification marker feature corresponds to the metabolite product time of the corresponding model of the metabolomics variable analysis preprocessing data. If so, the multidimensional verification result is successful, otherwise, the multidimensional verification result fails.

Images