TranslAGE
Platform for systematic, reproducible validation of epigenetic aging biomarkers (DNA-methylation clocks) across populations. Provides harmonized methylation datasets and precomputed clock scores, and introduces the composite STAR score benchmarking biomarkers on Stability (to technical/biological noise), Treatment response (to interventions/exposures), Associations (with clinical phenotypes), and Risk predictions (disease onset and mortality).
Composite
63.3
Experimental validation
N/A — retrospective validation platform over harmonized human methylation cohorts
Stages
Disease modelingTarget ID
Modalities
small molecule
Task types
regressionclassification
Size
clocks_scored: multiple standardized epigenetic clocks
evaluation_domains: 4
splits: {'train': 0, 'val': 0, 'test': 0}
note: Harmonized DNA-methylation cohorts with precomputed clock scores; STAR = Stability/Treatment-response/Associations/Risk; exact cohort N per release — see bioRxiv 10.64898/2026.05.11.723515
evaluation_domains: 4
splits: {'train': 0, 'val': 0, 'test': 0}
note: Harmonized DNA-methylation cohorts with precomputed clock scores; STAR = Stability/Treatment-response/Associations/Risk; exact cohort N per release — see bioRxiv 10.64898/2026.05.11.723515
License
Other — see bioRxiv preprint
First release
2026-05-11
Last updated
2026-05-11
Official site
Leaderboard
→ leaderboard
Dataset
Code / GitHub
→ repository
HuggingFace
→ HF
Paper
TranslAGE: A Comprehensive Platform for Systematic Validation of Epigenetic Aging Biomarkers · TranslAGE authors — see bioRxiv 10.64898/2026.05.11.723515 · 2026 · paper · doi:N/A — bioRxiv 10.64898/2026.05.11.723515 · 0 citations
Flags
none
Experts
—
Groups
—
Hosted by
—
Related benchmarks
Rubric (7-criterion)
rigor
4
coverage
4
maintenance
3
adoption
1
quality
4
accessibility
3
industry_relevance
3
Notes
Addresses a genuine reproducibility gap in aging-clock benchmarking by harmonizing methylation cohorts and standardizing clock scoring (rigor 4, quality 4). STAR is a differentiated, decision-relevant framework (stability, treatment response, phenotype associations, risk prediction) rather than MAE-to-chronological-age alone (coverage 4). Adoption 1 as a mid-2026 preprint; accessibility 3 pending public data/code release. Directly relevant to longevity biomarker qualification and intervention readouts.