Application of Mendelian randomization and bioinformatic analysis to construct a prognostic model for thyroid cancer and perform pan-cancer analysis

Abstract Objective: This study aimed to identify causal effects and potential molecular mechanisms of genes associated with THCA development. Methods: Bioinformatic analyses were performed to identify differentially expressed genes (DEGs) associated with THCA. Subsequently, Mendelian randomization (MR) analysis was conducted using large-scale eQTL data and THCA GWAS summary statistics to screen for candidate genes. The intersection of DEGs and MR-derived candidate genes was used to determine DEGs with potential causal associations with thyroid carcinogenesis. Functional enrichment analysis, pathway analysis, and