Screening and Identification of Key Biomarkers in Gastric Cancer: Three Genes Jointly Predict Gastric Cancer

Abstract

Background: Gastric cancer (GC) is the fifth most common cancer in the world and the second leading cause of cancer death. Gastric cancer screening is one of the effective strategies to reduce the mortality. If we use gastric cancer biomarkers with good clinical effect instead of gastroscopy screening, it can achieve early detection, early diagnosis, early treatment. Methods: We download four gene expression profiling datasets of gastric cancer (GSE118916, GSE54129, GSE103236, GSE112369), which obtained from the Gene Expression Omnibus (GEO) database. The DEGs between gastric cancer and adjacent normal tissues were detect to explore genes that may play a key role in gastric cancer. GO and KEGG analyses of overlap genes were performed by the Metascape online database, the protein-protein interaction (PPI) network was analyzed by the STRING online database, and we obtained the hub genes of PPI network via using the Cytoscape software. The survival curve analysis was performed by km-plotter and the stage plots of hub genes was performed by the GEPIA online database. We selected CDH3, LEF1 and MMP7 as candidate biomarkers to construct a back propagation neural network model. Results: This study shows that MMP7, CDH3 and LEF1 are highly expressed in gastric cancer tissues. Conclusions: The joint prediction of the MMP7, CDH3 and LEF1 is helpful for the early diagnosis of cancer.