Stomach cancer prognostic marker screening and classifying method based on gene expression profile

Abstract

The invention discloses a stomach cancer prognostic marker screening and classifying method based on a gene expression profile. The stomach cancer prognostic marker screening and classifying method comprises that a gene the expression of which is changed is screened out from data of the gene expression profile Cox proportional hazard model analysis is established aiming at the screened gene, a gene which has statistical significance is selected as a seed gene, a maximum likelihood model is constructed by means of clinical follow-up information data of the patient to screen out prognosis key genes, then the risk coefficient of each prognosis key gene in the patient is calculated, classification and statistic verification are carried out in dependence on the number of prognosis key genes of the patient, and an optimal classification mode is selected. According to the invention, dimensionality of high-dimension miscellaneous gene expression profile data can be effectively reduced, several key genes which can be applied to clinical detection easily are screened out from tens of thousands of genes, survival and the recurrence risk of the patient are predicted by means of the expression case of the several key genes.

Description

technical field [0001] The invention relates to the technical field of biological research, in particular to a method for screening and classifying gastric cancer prognosis markers based on gene expression profiles. Background technique [0002] Gastric cancer is one of the most common malignant tumors with high morbidity and mortality. Early diagnosis, reasonable assessment of prognosis and timely and appropriate intervention are very important. Current research reveals that gastric cancer is a genetic disease, which is an evolution process involving the participation of various oncogenes and anti-cancer genes, multi-stage and multi-pathway coordination, so that the gastric mucosa gradually develops into precancerous lesions, and then develops into gastric cancer. There are characteristic differentially expressed genes in the process from normal gastric mucosa to precancerous change and from precancerous change to gastric cancer. Molecular pathology combines molecular hybri...

AI Technical Summary

Problems solved by technology

At present, most studies on tumor classification methods based on gene expression profiles focus on two aspects: 1. Since gene expression profile data has the remarkable characteristics of high-dimensional samples, and there are a large number of redundant genes and noise in it, how to extract Disease-causing genes are still a difficulty; Second, the accuracy of gene expression profile data analysis has not yet reached the application level, and the final diagnosis of tumors, such as gastric cancer, still depends on medical experts

Images