Exploring the association between DNA methylation and pancreatic cancer susceptibility through epigenome-wide Mendelian Randomization and Multi-omics data integration

In this study, we employed cis-meQTL data to perform an EWMR analysis on 420,509 CpG sites, aiming to identify blood-based CpG sites linked to PC. To verify our key findings, we conducted sensitivity analyses, conditional Bayesian colocalization, replication studies, and meta-analysis. We then carried out chromatin state, functional, and pathway enrichment analyses for the CpG sites identified through EWMR as being associated with PC. Additionally, epigenome-wide association study (EWAS) and phenome-wide association study (PheWAS) were employed to explore whether these CpG sites were linked to other diseases or phenotypes. We also evaluated functional and clinical significance of the identified CpG sites by using druggable gene target analysis. Finally, we used multivariable Mendelian Randomization (MVMR) to conduct mediation analysis, incorporating multiple dimensions of expression quantitative trait loci (eQTLs).