Integrated bioinformatics and experimental validation reveal that CENPF activates mTORC1 signaling to promote colorectal cancer progression

The incidence and mortality rates of colorectal cancer (CRC) continue to increase, underscoring the need to identify key gene networks and clinically relevant biomarkers to better understand its molecular basis. This study aimed to systematically screen and validate centromere protein F (CENPF) as a core gene associated with CRC. In this study, using the GSE74602 dataset from the Gene Expression Omnibus (GEO), we identified differentially expressed genes (DEGs) with a false discovery rate (FDR) < 0.05 and |log2FC| > 1. Weighted gene co-expression network analysis (WGCNA) revealed that the blue and turquoise modules were significantly associated with CRC. The overlap of DEGs and module genes was imported into the STRING database to construct a protein–protein interaction (PPI) network, and 39 hub genes were subsequently identified using Cytoscape software. Machine learning approaches, including least absolute shrinkage and selection operator (LASSO) regression and random forest (RF), prioritized CENPF and topoisomerase II alpha (TOP2A) as candidate genes. Given that TOP2A has been extensively investigated in CRC, we focused on CENPF for further analyses. CENPF overexpression was validated across multiple datasets (GEPIA, GSE113513, and GSE44076). Receiver operating characteristic (ROC) analyses, including cross-validated evaluation, suggested strong diagnostic discrimination, and survival analysis indicated a possible association between high CENPF expression and poorer outcomes. Functional experiments demonstrated that silencing CENPF suppressed CRC cell proliferation, invasion, and migration. Mechanistically, CENPF promoted CRC progression through activation of mechanistic target of rapamycin complex 1 (mTORC1) signaling pathway. Overall, these findings indicate that CENPF may serve as a candidate diagnostic marker and potential therapeutic target in CRC.