Co-expression network analysis illustrates the importance of driver genes in colorectal cancer diagnosis, prevention, and therapy

Abstract

Abstract Background As the third most common form of cancer worldwide, colorectal cancer (CRC) is a major health concern. The overall aim of this study is to reconstruct a network in order to identify novel biomarkers for diagnostic use, prospective Endocrine Disrupting Chemicals (EDCs) for preventative use, and novel medications for therapeutic use in early-stage CRC. Material and Methods The driver genes linked with early-stage CRC were selected from the gene expression omnibus (GEO) and DriverDB databases. Then with the help of WGCNA (Weighted gene co-expression network analysis), the R package, the co-expression network was reconstructed. Following that, modules were chosen for further analysis. The possible biomarkers and hub genes were identified using the Cytoscape software and the cancer genome atlas (TCGA) database for diagnostic purposes. Then probable EDCs were identified using the Comptox database and the EDC-GENE network was reconstructed and the EDCs with a high degree of risk for preventive purposes were identified. As a next step, the drug-gene network was reconstructed to find effective drugs for colorectal cancer in its early stages. Results The co-expression network was constructed using the 1108 driver genes mRNA expression values of 70 early-stage CRC and 12 healthy control samples. The clustering results show that the overlapping gene set is divided into 27 modules. In our study, five modules (indicated by the colors of dark green, dark orange, light cyan, royal blue, and purple) were identified according to the average linkage hierarchical clustering and Zsummary less than 2. Then we find 17 high-degree genes of these modules as potential biomarkers for diagnostic issues. Moreover, we explored 25 potential high degrees of Endocrine Disrupting Chemicals that affect the main genes of each module for preventing purposes. Finally, we identified 27 potential high-degree drugs that affect the main genes of each module as treating aims. Then, these biomarkers, EDCs, and drugs that may be tested as a basis for future research were introduced. Conclusion The goal of this study was to identify candidate biomarkers for early detection, possible EDCs for prevention, and treatment agents for colorectal cancer. These biomarkers, EDCs, and drugs will help in the early detection, prevention, and treatment of colorectal cancer. Bioinformatics, computational biology, and systems biology methods were used to reach these claims; hence, they need to be tested in the lab. We anticipate that these results will provide important new insights into the etiology and early evolution of CRC and that they will inspire the development of novel approaches to treating this aggressive and lethal malignancy.