Multi-organ metabolome biological age implicates cardiometabolic conditions and mortality risk

Biological aging clocks across organs and omics data, including clinical phenotypes, neuroimaging, proteomics, and epigenetics, have proven instrumental in advancing our understanding of human aging and disease. Here, we expand this aging clock framework to plasma metabolomics by developing 5 organ-specific metabolome-based biological age gaps (MetBAGs) using 107 plasma non-derived metabolites from 274,247 UK Biobank participants. Our multi-organ MetBAGs were trained using Lasso regression and neural networks, achieving a mean absolute error of approximately 6 years (0.25