Integrating single-cell sequencing and Mendelian randomization to explore molecular mechanisms in pancreatic cancer

Pancreatic cancer, a highly aggressive malignancy originating from pancreatic ductal epithelial cells, is associated with dysregulation of key genes and signaling pathways. Recent advances in research on the tumor microenvironment, particularly immune infiltration, have provided novel therapeutic insights. This study aimed to identify critical genes and signaling pathways involved in pancreatic cancer, offering potential directions for mechanistic exploration and therapeutic development. Utilizing data from the TCGA, GEO, eQTLGen, and EBI databases, we integrated single-cell sequencing with Mendelian randomization (MR) to identify key pancreatic cancer-associated genes. Functional enrichment analyses (GSEA and GSVA) were performed to elucidate pathway involvement, whereas immune infiltration analysis was used to evaluate interactions between key genes and immune regulation. A prognostic nomogram model was constructed, and pseudotime analysis was conducted to track dynamic gene expression during cellular development. Three key genes— CTSH , MUC1 , and SLBP —were identified. GSEA revealed enrichment of CTSH in intestinal immune networks for IgA production, NF-κB signaling, and B-cell receptor pathways. MUC1 is associated with oxidative phosphorylation and retinol metabolism, whereas SLBP is linked to nucleotide metabolism and p53 signaling. GSVA highlighted CTSH in the reactive oxygen species and p53 pathways, MUC1 in xenobiotic metabolism, and SLBP in E2F targets and G2/M checkpoint regulation. These genes exhibited high activity in epithelial‒mesenchymal transition, MYC targets, and coagulation pathways. Immune infiltration analysis demonstrated significant correlations between key genes and immune cell subtypes (e.g., CTSH with naive B cells and M1 macrophages). The nomogram demonstrated moderate predictive performance for patient survival, yielding AUC values of 0.625 at 1 year and 0.749 at 3 years. This study identified CTSH , MUC1 , and SLBP as candidate genes associated with pancreatic cancer, with potential relevance to immune, metabolic, and cellular regulatory processes. These findings provide a basis for further investigation of their biological and prognostic relevance.