Method of inferring gene pathway activity

Abstract

The invention provides a method of inferring gene pathway activity. The method comprises: obtaining samples and corresponding pathway networks and expression values of all genes thereof, and using t values of a gene t test and Pearson correlation coefficients as weights of the genes for weighting processing on the expression values of all the genes; obtaining types of interaction among genes in the same pathways and corresponding intensity thereof according to topology structures of the pathway networks of all the samples, and using the intensity of the types of interaction among the genes andthe weighted expression values of all the genes to obtain expression values of interaction among all the genes; and integrating the expression values and the interaction expression values of all thegenes, using a principal-component-analysis (PCA) method for analysis, and further defining all obtained first principal components as activity scores of the pathways. By implementing the method, importance of the genes and importance of interaction among the genes are simultaneously considered for inferring the activity of the pathways, and thus evaluation on biological pathway sample status is realized.

Description

technical field [0001] The invention relates to the technical field of gene detection, in particular to a method for deducing gene pathway activity. Background technique [0002] Many recent research methods propose to find more robust biological markers at the functional level to break through the instability of single gene markers. Because genes do not participate in biological processes alone, gene products usually act synergistically in the form of functional modules or signal cascades. Functional modules that are dysregulated at a high level may be more stable as biomarkers than single genes, and various noises have little influence on them . Functional biological markers can effectively reduce tissue heterogeneity and sample genetic heterogeneity, and at the same time effectively analyze the relationship between important functional pathways and diseases. Therefore, integrating the expression profiles of functionally related genes and extracting classification featur...

AI Technical Summary

Problems solved by technology

[0004] However, the above-mentioned pathway activity inference method only utilizes the significant genes in the pathway, and does not consider the interaction information between genes. It only regards the pathway as a simple collection of individual genes, but ignores the internal gene topology of the pathway network. Structural information, which loses a lot of important information about gene-to-gene communication

Images