A genetic map of human metabolism across the allele frequency spectrum

Genetic studies of human metabolism identified unknown disease processes and novel metabolic regulators, but have been limited in scale and allelic breadth. Here, we provide a data–driven map of the genetic regulation of circulating small molecules and lipoprotein characteristics (249 metabolic traits) measured using protein nuclear magnetic resonance spectroscopy (1H–NMR) across the allele frequency spectrum in ~450.000 individuals. In trans–ancestry analyses, we identify 29,824 locus–metabolite associations mapping to 753 regions with effects largely consistent between men and women and major ancestral groups represented in UK Biobank. We develop a framework for classifying the observed extreme genetic pleiotropy, enabling identification of upstream master regulators of lipid metabolism (proportional pleiotropy), such as ANGPTL3. We establish rare–to–common allelic series by integrating machine–learning guided effector gene assignments with rare exonic variant analyses providing high confidence gene assignments at >100 loci, including less established regulators of lipid metabolism like SIDT2. At 17 such loci we observed phenotypic heterogeneity among variants mapping to the same gene indicating differential metabolic roles of the altered gene product. We identify VEGFA as a potential modulator of HDL–mediated risk for coronary artery disease. Our results demonstrate how rare–to–common genetic variation combined with deep molecular profiling can identify unknown and inform on poorly understood regulators of human metabolism to guide prevention and treatment of diseases.