DNA-based clock estimates age in East Asian populations
Estimates based on chemical marks at 38 DNA sites differed from participants’ actual ages by 6.22 years on average in the test group.
Limit: Testing split pooled datasets instead of using a separately recruited validation cohort
Edited by Boris Djordjevic · AgingResearch.News
In a computational modeling study, researchers analyzed whole-blood DNA data from 3,637 people of East Asian ancestry. They used 2,546 people aged 6.40–84.00 years to build the model and 1,091 aged 1.34–85.65 years to test it.
The model, called the EAS Clock, used DNA methylation, chemical marks on DNA, at 38 sites to estimate chronological age—how long someone has lived. Estimates tracked actual ages in both groups. The mean absolute error, the average size of a prediction’s error regardless of direction, was 7.48 years in training and 6.22 years in testing. Overestimates and underestimates largely balanced out in both groups.
Why it matters
Age-related disease highlights the need for molecular tools to assess aging. Such tools need evaluation across ancestries because clocks developed mainly in non-East Asian populations may be less well calibrated elsewhere.
Caveats
Testing used a random split of pooled datasets, not a separately recruited validation cohort. Predicting chronological age does not establish that the clock measures biological aging or predicts disease risk.
The paper
An epigenetic clock for chronological age estimation in East Asian populations
National Taiwan University
NAR Genomics and Bioinformatics, Jun 30, 2026, Open access
Comments
Be the first to comment.