Novel method for chromatography-mass spectrometry metabolomics data analysis

Abstract

The present invention relates to a novel method for chromatography-mass spectrometry metabolomics data analysis, and belongs to the technical field of analytical chemistry. Based on chromatography-mass spectrometry data characteristics, the method establishes a metabolite mass spectrometry database XMet; then the XMet is projected to a statistical analysis model established by samples in the study system, so as to obtain a metabolite projection score matrix TMet; first two columns of the TMet are plotted to obtain metabolites with differentiation; and metabolites with differentiation can be identified through retrieval of the XMet and the mass spectrometry database. Based on the former methods, the invention sets up the novel method for chromatography-mass spectrometry metabolomics data analysis, so as to conveniently screen and directly identify the metabolites with differentiation, thereby effectively solving the problems in the existing gas chromatography-mass spectrometry metabolomics data analysis strategy.

Description

technical field [0001] The invention relates to a chromatography-mass spectrometry metabolomics data analysis method, which belongs to the technical field of analytical chemistry. Background technique [0002] Metabolomics is an emerging technology in the field of systems biology, and chromatography-mass spectrometry (such as GC-MS by gas chromatography-mass spectrometry, LC-MS by liquid chromatography-mass spectrometry, etc.) is the mainstream analytical technique in metabolomics research. Chromatography-mass spectrometry instruments often generate a large amount of data. How to effectively analyze these data to obtain the metabolic profile characteristics of the research object is the key technology and bottleneck problem in current research in this field. [0003] At present, the common strategy for studying chromatographic-mass spectrometry metabolomics data is to use the mass spectrum fragmentation peak intensity at the retention time Rt-mass-to-charge ratio m / z as a va...

AI Technical Summary

Problems solved by technology

However, this data analysis strategy has the following problems: (1) With this strategy, each sample can generate as many as thousands of variables (intensities of mass spectrum fragmentation peaks), while the number of samples in metabolomics research is generally dozens , which will lead to an imbalance between the number of variables and the number of samples, which will cause problems for subsequent analysis; (2) This strategy focuses on the information of mass spectrum peaks, and mass spectrum peaks include addition peaks, isotope peaks, etc. in addition to fragment ion peaks. Therefore, there is about 90% redundant information in the above-mentioned large number of variables; moreover, the variables are generally intensity information recorded at a certain mass-to-charge ratio and retention time, all of which complicate the qualitative identification of differential metabolites in the later stage; ( 3) Due to the large number of endogenous metabolites in metabolomics research, the content varies, and sometimes there may be phenomena such as chromatographic peak overlap and deformation, which will bring errors to the data processing process and affect the reliability of subsequent statistical analysis results

However, the MPA method is established for NMR data, and its necessary condition is to obtain the NMR spectrum information of metabolites; and biological samples generally contain hundreds or thousands of metabolites, so it is difficult to collect spectral information of so many metabolites. hard to do

Images