What this means
Transcriptomic biological age clocks may track age-related transcriptional variation in skeletal muscle, but it remains unclear whether they detect sarcopenia-related molecular variation beyond chronological age. This multicohort study benchmarked directly implementable RNA-based age models in public human skeletal muscle RNA-seq datasets.
A plain-English summary is on its way. Until then, this is the start of the authors' own summary.
- Who it was tested in
- Not stated
- Kind of research
- A study
The science
- Original title
- RNA-Based Biological Age Clocks Track Chronological Age in Skeletal Muscle but Show Limited Discrimination of Sarcopenia.
- Authors
- Ledesma-Gallardo SD, López-Espinosa E, Ortega-Meléndez AI et al.
- Journal
- Cell biochemistry and biophysics
- Published
- Oct 24, 2026
- Study type
- Study
- PMID
- 42925470 ↗
Abstract
Transcriptomic biological age clocks may track age-related transcriptional variation in skeletal muscle, but it remains unclear whether they detect sarcopenia-related molecular variation beyond chronological age. This multicohort study benchmarked directly implementable RNA-based age models in public human skeletal muscle RNA-seq datasets. GSE164471 was used as a healthy-aging reference, GSE111006/GSE111010 as a pooled discovery benchmark, GSE167186 as external validation, and GSE174106 as biological validation using appendicular lean mass trajectories. RNAAgeCalc skeletal muscle and across-tissue models were implemented for age estimation and sarcopenia discrimination. Gene coverage was used only as a preliminary technical-compatibility assessment, after which the published SkeletAge model was directly applied to GSE111006 and GSE111010 for independent chronological-age prediction and sarcopenia-discrimination analyses; GSE164471, which contributed to SkeletAge model development, was used only for an in-sample implementation check. Both RNAAgeCalc models showed high gene coverage. In GSE164471, the skeletal muscle model showed stronger chronological-age tracking than the across-tissue model (r = 0.604 vs. 0.266; MAE 15.2 vs. 16.7 years), but retained substantial absolute prediction error and positive mean bias (+ 10.4 years), indicating that stronger correlation did not imply adequate calibration. In the discovery benchmark, between-group differences in chronological age, transcriptomic age, and age-acceleration residuals were not statistically significant, with small-to-moderate standardized effect estimates and confidence intervals spanning zero. Logistic discrimination was influenced mainly by cohort origin, while transcriptomic aging measures provided limited independent discrimination. External validation showed broad age-group structure with substantial overlap between old healthy and sarcopenic participants, and no differences were detected across appendicular lean mass trajectories. Current RNAAgeCalc implementations therefore appear to track chronological-age-related transcriptomic structure more effectively than sarcopenia-related variation, although smaller disease-related effects cannot be excluded, and these models should not currently be considered standalone sarcopenia biomarkers.
Ledesma-Gallardo SD, López-Espinosa E, Ortega-Meléndez AI et al. RNA-Based Biological Age Clocks Track Chronological Age in Skeletal Muscle but Show Limited Discrimination of Sarcopenia. Cell biochemistry and biophysics. 2026. doi:10.1007/s12013-026-02200-2. PMID: 42925470. https://pubmed.ncbi.nlm.nih.gov/42925470/