Abstract
BackgroundMetabolites are closely linked to individual health and disease conditions. Identifying genetic factors influencing metabolite levels in specific diseases can enhance our understanding of disease etiology and informing precision medicine. This study aims to characterize the genetic architecture of metabolites in specific disease states and explore their potential biological functions.MethodsWe conducted a comprehensive genome-wide metabolite quantitative trait locus (metQTL) analysis of 249 plasma metabolites across 40 disease states, based on nuclear magnetic resonance (NMR) data. To predict the biological significance of metQTLs, we performed a systematic functional annotation encompassing the analysis of genomic positions, heritability assessment, histone and transcription factor (TF) enrichment, effector gene identification, and potential drug targets evaluation. Furthermore, Mendelian randomization (MR) analyses were applied to uncover causal metabolites associated with diseases, and polygenic risk score (PRS) models were constructed to assess their predictive capacity for disease outcomes.ResultsAcross 40 common disease types, we identified 283,563 metQTL-metabolite association pairs involving 249 metabolites and 149,984 metQTLs derived from 26,536 independent loci. Functional annotations indicated that these metQTLs influence chromatin activity and transcription factor binding, suggesting their key roles in epigenetic regulation. Mendelian randomization analysis revealed 104 reliable causal evidence between metabolites and diseases. Additionally, metQTL-derived disease PRS models demonstrated excellent performance in the risk stratification of 8 diseases, offering a framework for translating genetic resources into clinical applications. An online platform, “metQTL-Atlas” (https://metqtl.whu.edu.cn/home), has been established for convenient browsing and downloading of our comprehensive results.ConclusionsThis study provides a comprehensive resource that delineates the genetic architecture of metabolites across diverse disease contexts, offering new insights into disease etiology and advancing precision medicine through enhanced risk prediction and therapeutic target discovery.