Identification of significant genes associated with prognosis of gastric cancer by bioinformatics analysis.

Background: Gastric cancer (GC) ranks second in mortality among all malignant diseases worldwide. However, the cause and molecular mechanism underlying gastric cancer are not clear. Here, we used integrated bioinformatics to identify possible key genes and reveal the pathogenesis and prognosis of gastric cancer.

Methods: The gene expression profiles of GSE118916, GSE79973, and GSE29272 were available from the Gene Expression Omnibus (GEO) database. Differentially expressed genes (DEGs) between GC and normal gastric tissues were screened by R software and Venn diagram software. GO and KEGG pathway enrichment of DEGs was performed using the DAVID database. A protein-protein interaction (PPI) network was established by STRING and visualized using Cytoscape software. Then the influence of hub genes on expression and survival was assessed using TCGA database.

Results: A total of 83 DEGs were found in the three datasets, including 41 up-regulated genes and 42 down-regulated genes. These DEGs were mainly enriched in extracellular matrix organization and cell adhesion. The enriched pathways obtained in the KEGG pathway analysis were extracellular matrix (ECM)-receptor interaction and focal adhesion. A PPI network of DEGs was analyzed using the Molecular Complex Detection (MCODE) app of Cytoscape. Four genes were considered hub genes, including COL5A1, FBN1, SPARC, and LUM. Among them, LUM was found to have a significantly worse prognosis based on TCGA database.

Conclusions: We screened DEGs associated with GC by integrated bioinformatics analysis and found one potential biomarker that may be involved in the progress of GC. This hub gene may serve as a guide for further molecular biological experiments.