Combined bioinformatics technology to explore pivot genes and related clinical prognosis in the development of gastric cancer

Abstract

Abstract To screen the key genes in the development of gastric cancer and their influence on prognosis. The GEO database was used to screen gastric cancer-related gene chips as a training set, and the R packages limma tool was used to analyze the differential genes expressed in gastric cancer tissues compared to normal tissues, and then the selected genes were verified in the validation set. The String database was used to calculate their Protein–protein interaction (PPI) network, using Cytoscape software's Centiscape and other plug-ins to analyze key genes in the PPI network. The DAVID database was used to enrich and annotate gene functions of differential genes and PPI key module genes, and further explore correlation between expression level and clinical stage and prognosis. Based on clinical data and patient samples, differential expression of key node genes was verified by immunohistochemistry. The 63 characteristic differential genes screened had good discrimination between gastric cancer and normal tissues, and are mainly involved in regulating extracellular matrix receptor interactions and the PI3k-AKT signaling pathway. Key nodes in the PPI network regulate tumor proliferation and metastasis. Analysis of the expression levels of key node genes found that relative to normal tissues, the expression of ITGB1 and COL1A2 was significantly increased in gastric cancer tissues, and patients with late clinical stages of tumors had higher expression of ITGB1 and COL1A2 in tumor tissues, and their survival time was longer (P < 0.05). This study found that ITGB1 and COL1A2 are key genes in the development of gastric cancer and can be used as prognostic markers and potential new targets for gastric cancer.