Abstract
Genomics and bioinformatics methods were used to screen genes and molecular markers correlated with colorectal cancer incidence and progression, and their biological functions were analyzed. Differentially expressed genes were obtained using the GEO2R program following colorectal cancer chip data GSE44076 retrieval from the Gene Expression Omnibus gene expression comprehensive database. An online database (David) that combines annotation, visualization, and gene discovery was utilized for investigating genes. Pathway and protein analyses were performed via resources from the Gene Ontology (GO) and the Kyoto Encyclopedia of Genes and Genomes (KEGG). Visual analysis of the KEGG pathway was carried out according to ClueGO and CluePedia to establish the PPI network of gene interaction between pathways; the genes with the highest connectivity were screened by the molecular complex detection analysis method as Hub genes in this study; gene expression was verified by GEPIA online analysis tool, and Kaplan-Meier survival curve was drawn for prognosis analysis. By analyzing GSE44076 microarray data, 86 genes were selected, and colorectal cancer tissues' upregulation was observed in 27 genes and downregulation in 59 ones. GO assessment revealed that the differentially expressed genes were basically correlated with retinol dehydrogenase activity, carbon dehydrogenase activity, collagen-containing extracellular matrix, anchored component of memory, and cellular hormone metabolic process. Moreover, the KEGG assessment revealed that the differential genes contained various signal pathways such as retinol metabolism, chemical carotenogenesis, and nitrogen metabolism. Through further analysis of the PPI protein network, 4 clusters were obtained, and 16 Hub genes were screened out by combining the degree of each gene. Through the analysis of each gene on the prognosis of colon cancer through the GEPIA online analysis website, it was found that the expression levels of AQP8, CXCL8, and ZG16 genes were remarkably associated with colon cancer prognosis (P < 0.05). Genomics and bioinformatics methods can effectively analyze the genes and molecular markers correlated with colorectal cancer incidence and progression, help to systematically clarify the molecular mechanism of 16 key genes in colorectal cancer development and progression, and provide a theoretically valid insight for the screening of diagnostic markers of colorectal cancer and the selection of accurate targets for drug therapy.