Gene activity clock tracks relative age across human tissues
Trained on 17,212 healthy-donor samples, the model predicted relative age across tissues and several methods of measuring gene activity.
Our read
Biomarkers- Studied:
- Computational study of healthy-donor gene-activity samples
- What changed:
- Pasta accurately predicted relative age across diverse human tissues and methods of measuring gene activity
- The catch:
- The clock scores gene-activity states, not lifespan or health
- In the record:
- The 5th paper on piperlongumine with results in people; 22 on record in all.
Editor of record: Boris Djordjevic · AgingResearch.News
Human gene-activity data from 17,212 healthy-donor samples across 21 studies underpinned Pasta, a computational aging model. Researchers trained it to distinguish older from younger samples using pairs separated by at least 40 years.
Pasta predicted relative age across tissues and measurement methods, including datasets measuring gene activity in individual cells. Its scores aligned with senescence, a state in which cells stop dividing, and with stem-like cellular states. Applied to more than three million gene-activity profiles from 248 cell lines, it identified chemical and genetic changes associated with older or younger cellular states. Follow-up experiments in human cancer cell lines found that one predicted compound, pralatrexate, induced senescence.
Why it matters
Measuring age-related gene activity across tissues could help researchers compare cellular aging and screen chemical and genetic changes for further testing.
Caveats
Cellular age scores describe gene-activity states, not demonstrated gains in lifespan or health. The compound experiments used human cancer cell lines, not people.
The paper
Stockholm University
Advanced Science, Jul 27, 2026, Open access
Comments
Be the first to comment.