MetabHub maps links between blood molecules and disease risk
Data from 364,795 European UK Biobank participants yielded 2,950,764 associations across health measurements, environmental exposures, genetic variants, blood molecules and diseases.
Edited by Boris Djordjevic · AgingResearch.News
In an observational study, researchers analyzed data from 364,795 European UK Biobank participants to build MetabHub, an online resource about blood chemistry and future disease risk. It included 251 traits measured in plasma, the liquid part of blood, alongside clinical measurements, environmental exposures, genetic variants and 57 diseases that developed during follow-up. It mapped associations for both sexes together and separately.
The website also offered statistical mediation analyses, which examine whether an intermediate factor may help account for an association. Researchers built blood-based disease prediction models and evaluated their risk scores over about 14 years of follow-up. Users can upload their own profiles to apply the pre-trained models.
Why it matters
Long-term disease risk is relevant to efforts to preserve health in later life. The authors present the resource as a tool for exploring disease mechanisms and tailoring prevention.
Caveats
Observational associations and statistical mediation analyses do not establish cause and effect. The data came from European UK Biobank participants, and the abstract does not report prediction accuracy or testing in other populations.
The paper
Nanjing Medical University Tianyuan Honors School; First Affiliated Hospital of Nanjing Medical University
Nucleic Acids Research, Oct 29, 2026
