Identification of prognostic-associated genes in colorectal cancer based on genome-wide expression profiles

Abstract Background: Colorectal cancer (CRC), a severe disease in the world. The dysregulation of genes plays a significant role in cancer. Method: In the present study, we first identified differentially expressed genes (DEGs) of CRC in the GSE71187 data set from the Gene Expression Omnibus (GEO) database. And through GO terminology and Gene and Genome (KEGG) pathway, the up-regulated DEG and down-regulated genes were enriched. Then, we selected hub genes and studied their related prognostic value through protein-protein interaction (PPI) network and integrated bioinformatics analysis. Finally, 7 genes with prognostic value were obtained by survival analysis and multivariate Cox proportional hazards regression models, as well as verified in the Oncomine and UALCAN database.Result: A total of 3,588 DEGs were identified with 1,474 upregulated and 3,114 downregulated. Then, GO terms and Genes and Genomes (KEGG) pathway analysis revealed that upregulated DEGs and down-regulated genes were mainly involved in important GO terms and KEGG pathway, for example, defense response to other organism, leukocyte differentiation and epidermis developmen. In the PP network analysis, 6 hub modules and 86 module genes were identified. In the end, 7 genes with prognostic value were obtained through survival and multi-factor analysis. They have significant differences in varying degrees at the level of DNA, mRNA and protein.Conclusion: The results of the study may provide novel insight into the genes and associated pathways involved in CRC, and aid the identifcation of novel therapeutic targets and prognostic marker of CRC.