Proposed entropy-based DNA methylation clock tested in benchmark cohorts
The abstract describes a proposed DNA methylation clock, Stacy-GB, that uses methylation entropy plus machine learning; it reports a mean absolute error of 3.74 years in benchmark cohorts and links its age acceleration to several clinical conditions. [1]
Automatic summary · not editorially reviewed
The key findings
The proposed Stacy-GB clock combines CpG-level methylation entropy from a Generalized Gamma Distribution with a LightGBM regressor. [1]
In the ComputAgeBench benchmark across independent cohorts, Stacy-GB had a mean absolute error of 3.74 years and a median error of 2.41 years. [1]
Age acceleration estimated by Stacy-GB was associated with ischemic heart disease, HIV infection, multiple sclerosis and Werner syndrome. [1]
The source at a glance
- Study type / format
- Model development and benchmark evaluation (abstract)
- Population / context
- People in independent cohorts evaluated with the ComputAgeBench epigenetic clock benchmark, as described in the abstract.
- Summary based on
- Study abstract
- Publication status
- Preprint · peer-review status not confirmed
- Original publication
- August 20, 2026
Methylation entropy as clock input
The authors note that common epigenetic clocks are mostly empirical models that do not explicitly describe methylation variability or connect directly to physical aging mechanisms. They propose an information-theoretic framework: at each CpG site they model the population distribution of methylation β-values with a three-parameter Generalized Gamma Distribution and derive a closed-form expression for its differential Shannon entropy. This CpG-level entropy describes methylation variability and serves as a criterion for selecting loci. The entropy representation is then combined with a LightGBM regressor in the proposed Stacy Gradient Boosting Clock (Stacy-GB).
Supporting source [1]
Reported error in benchmark cohorts
The model was evaluated across independent cohorts using the ComputAgeBench epigenetic clock benchmark. Stacy-GB reached a mean absolute error of 3.74 years and a median error (bias) of 2.41 years, which the authors describe as significantly outperforming state-of-the-art epigenetic clock baselines. The abstract does not report cohort sizes, follow-up, or the specific baseline clocks compared.
Supporting source [1]
Age acceleration and clinical pathologies
Age acceleration estimated by Stacy-GB was associated with several clinical pathologies: ischemic heart disease, HIV infection, multiple sclerosis and Werner syndrome. The abstract presents these as associations, not as evidence that the clock causes these conditions or predicts their course. The authors frame Stacy-GB as a potentially accurate and biophysically grounded tool for clinical aging research; this is a proposal, not a validated clinical product.
Supporting source [1]
How to interpret this
This is an abstract only; the full text was not reviewed. Cohort sizes, follow-up, the specific baseline clocks and the strength of the clinical associations are not reported here. The associations are not causal, and Stacy-GB is a proposed research tool, not a validated clinical product.
Based on the study abstract; the full paper, tables and supplementary data were not reviewed.
Source & supporting passages
Nasrolahpour H, Jandera A, Skovranek T, Despotovic V, Pellegrini M. · europepmc.org · August 20, 2026