Machine Learning Identification of TSPAN7 as a Key Target Linking Type 2 Diabetes Mellitus and Colorectal Cancer

Abstract

Background Type 2 Diabetes Mellitus (T2DM) and Colorectal Cancer (CRC) are significant global public health challenges with a notable epidemiological association. This study aims to explore the molecular mechanism behind this epidemiological association. Methods Weighted Gene Co-expression Network Analysis (WGCNA) and differential expression gene (DEG) analysis were conducted to identify shared genes between T2DM and CRC. Machine learning algorithms, including LASSO, Random Forest, and Support Vector Machine (SVM), were employed to identify hub genes. IOBR and clusterProfiler packages were used for immunoinfiltration assessment and enrichment analysis, respectively. Results We identified 27 shared genes between T2DM and CRC, with TSPAN7 emerging as a key hub gene linking the two conditions. TSPAN7 expression was significantly lower in disease groups compared to control groups across multiple cohorts, demonstrating excellent diagnostic accuracy. Enrichment analysis revealed involvement of these genes in various metabolic activities and pathways, including sulfur metabolism, selenium metabolism, renin secretion, pantothenate and CoA biosynthesis, TRP channel regulation, and efferocytosis. Conclusion This study provides new insights into the mechanisms underlying the association between T2DM and CRC by identifying TSPAN7 as a key target. The findings offer theoretical evidence for developing new diagnostic markers and therapeutic strategies for these diseases.