Carcinogenesis, Teratogenesis & Mutagenesis ›› 2019, Vol. 31 ›› Issue (4): 308-314.doi: 10.3969/j.issn.1004-616x.2019.04.008

Previous Articles     Next Articles

Investigation and evaluation of gastric cancer core genes using bioinformatics

CHEN Xiuqiong, MENG Fanqiao, XIONG Hua, WANG Yali, ZHOU Yangmei, TANG Wenhua, ZOU Yanmei   

  1. Cancer Center, Tongji Hospital Affiliated to Tongji Medical College, Huazhong University of Science and Technology, Wuhan 430030, Hubei, China
  • Received:2019-03-31 Revised:2019-06-28 Online:2019-07-30 Published:2019-08-23

Abstract: OBJECTIVE:Apply bioinformatics analyses to explore relationships between hub genes and diagnosis/prognosis of gastric cancer and to possibly enhance molecular diagnosis,targeted therapy and prognosis of gastric cancer. METHODS:Three gastric cancer-associated mRNA expression profiles were down-loaded from the GEO database. The Limma package of R software was used to screen out differentially expressed genes (DEGs) in microarrays. Then,Funrich software was employed to take the intersection of the three expression profiling chips,and further obtain DEGs in all three expression profiles. DAVID (Database for Annotation,Visualization and Integrated Discovery) database was utilized to do the GO function annotation and KEGG enrichment analysis for DEGs,STRING online website and Cytoscape software network analysis,plug-in CytoHubba were applied to construct protein interaction network (PPI) and visual analysis,and then screen out the core genes (Hub Gene). The relationship between core genes and prognosis of patients with gastric cancer was analyzed in KM database. The diagnostic value of core genes with prognostic significance was quantified and visualized by GraphPad software. Finally,Pearson method was used to examine correlations among core genes. RESULTS:In the three datasets,there were 1 839 DEGs,851 up-regulated genes,and 988 down-regulated genes. After screening out overlapping genes,there were 66 genes with significant differential expression in the three expression profiles,consisting of 24 up-regulated genes and 42 down-regulated genes. GO enrichment analyses show that the functions of DEGs were mainly concentrated in extracellular space,extracellular exosomes,digestion,extracellular matrix tissue,and collagen fibrous tissue. KEGG enrichment analyses show that involved pathways were mainly in protein digestion and absorption,gastric acid secretion,nitrogen metabolism,ECM-receptor interaction,mineral absorption. In the PPI network,Cytoscape visual analyses show that differential expression of the 10 core genes was closely related to the occurrence of gastric cancer. KM database searches identify that patients with low expression of FN1 and COL1A1 had better prognosis than those with higher expression. The AUC of FN1 and COL1A1 were 0.93 and 0.90,respectively,indicating that both had high diagnostic values. The correlation coefficient,R=0.59,was obtained through the Pearson Correlation analyses which suggest that the expression of COL1A1 and FN1 in gastric cancer was positively correlated. CONCLUSION:The data show that both FN1 and COL1A1 can potentially be important targets for prognosis,early diagnosis and development of targeted drugs for gastric cancer.

Key words: gastric cancer, bioinformatics, differentially expressed genes, prognosis, treatment, target

CLC Number: