Molecular Mechanisms and Clinical Predictive Value of Colorectal Cancer Susceptibility Genes: WGCNA Analysis and Mendelian Randomization Study

Background and objective: Colorectal cancer (CRC) is a highly heterogeneous disease, making treatment and prognosis prediction challenging. Early diagnosis of CRC and identification of gene expressions associated with its onset are crucial for prognosis, especially before clinical symptoms appear. This study aims to explore potential key genes involved in CRC and evaluate their clinical application in predicting the disease. Methods: This study utilizes differential expression analysis and Weighted Gene Co-expression Network Analysis (WGCNA) to identify novel susceptibility modules and key genes associated with colorectal cancer (CRC). Through KEGG and GO analyses, we aim to investigate the potential functions of these key genes. Subsequently, we will construct a Nomogram model and assess its diagnostic value for CRC using ROC curves. Based on genome-wide association studies, a Mendelian randomization analysis will be conducted to determine the causal relationship between these key genes and CRC. Finally, we will explore the association between these key genes, which are causally linked to CRC risk factors, and immune cell infiltration. Results: A gene co-expression network was constructed using WGCNA, from which key modules related to colorectal cancer (CRC) were identified, along with 963 overlapping key genes derived from WGCNA. GO and KEGG pathway enrichment analyses revealed that these genes are involved in the biosynthesis of ribonucleoprotein complexes, rRNA metabolic processes, chromatin organization-regulated signaling pathways, as well as cell cycle, DNA replication, and ribosome-related pathways. Using Cytoscape software, we identified the top five highly expressed genes: CDC2, CCNB1, CCNA2, TOP2A, and CCNB2. We then developed a Nomogram model, which effectively predicts the risk of CRC. The performance of this model in CRC diagnosis was further validated through ROC curve analysis, showing promising diagnostic accuracy. Finally, we focused on CDC2 and observed a causal relationship between CDC2 and immune cell infiltration in CDAD. Through inverse variance-weighted analysis, we found that CDC2 significantly increased the risk of CDAD, with an OR of 1.0005 (95% CI = 1.0001-1.001, P = 0.01). Conclusion: We successfully identified the core genes associated with colorectal cancer (CRC). This finding provides important insights for further research into early diagnostic methods for CRC, while also contributing to the understanding of the molecular mechanisms underlying CRC risk genes.