Abstract
BACKGROUND: Sepsis, a life-threatening syndrome triggered by a dysregulated host response to infection, continues to impose a substantial global health burden. Advances in genomics and transcriptomics now enable systematic exploration of sepsis pathogenesis at the genetic level. The integration of genome-wide association studies and transcriptome-wide association studies (TWAS) offers a powerful framework to identify causal genetic variants and delineate molecular mechanisms underlying sepsis susceptibility and clinical outcomes. METHODS: A cross-tissue TWAS was implemented using UTMOST to integrate sepsis genome-wide association studies summary statistics with transcriptomic data from the Genotype-Tissue Expression version 8 (GTEx v8) project. Candidate genes were validated through complementary approaches: FUSION, FOCUS, and MAGMA. Tissue-specific and pathway enrichment analyses were applied to prioritize sepsis-associated genes and characterize their functional roles in disease-relevant biological processes. Bayesian colocalization and two-sample Mendelian randomization (MR) analyses were employed to infer putative causal relationships between prioritized genes and sepsis risk. RESULTS: Four genes-ZCCHC4, PDGFB, C18orf54, and ATG4B-demonstrated significant associations with sepsis susceptibility in cross-tissue analyses. Two-sample MR provided evidence for causal effects of genetically regulated expression of these genes on sepsis risk. Bayesian colocalization identified shared causal variants between sepsis-associated loci and expression quantitative trait loci (eQTLs), implicating dysregulation of inflammatory and autophagy pathways in sepsis pathogenesis. CONCLUSIONS: Our results highlight the efficacy of cross-tissue TWAS in mapping sepsis-associated loci and elucidating the genetic architecture underlying sepsis susceptibility. These prioritized loci constitute compelling targets for functional validation and represent actionable candidates for therapeutic intervention in sepsis.