Identification of genes in ulcerative colitis associated colorectal cancer based on centrality analysis of co-expression network

Abstract

PreviousColorectal cancer (CRC) is awell-recognized complication of Ulcerative colitis (UC) and patients with UC have ahigher incidence of CRC than the general population. Early detection and mechanism of colitis-associated colorectal cancer (CAC) is still challenging. The aim of present study is to identify genes associated with CAC by centrality analysis of co-expression networks. Co-expression networks of CRC and UC were constructed by empirical Bayes approach based on top 200 gene signatures which identified by the model of genome-wide relative significance and genome-wide global significance across multiple datasets. Centrality of degree, stress centrality, betweenness centrality and closeness centrality of co-expression networks were selected to explore hub genes presented in CRC and UC. Validation of mRNA expression in CRC patients was conducted by real-time quantitative Polymerase Chain Reaction (qPCR). Pathway analysis was conducted based on Kyoto Encyclopedia of Genes and Genomes database. We found 21 common genes, such as SLC4A4 and AQP8, both existed in CRC and UC top 200 genes. By accessing centralities analyses of co-expression networks, HPGD and AQP8 were common hub genes in CRC and UC, and various centralities analyses of the same gene were not consistent. Patients with alteration of AQP8 have significantly reduced the survival rate according to real-time qPCR results. Our study displayed genes associated with CAC (AQP8 and HPGD), and they might be reliable biomarkers for early detection and therapies of CAC.