A recent study published in Aging-US details a new method for predicting age using DNA methylation patterns found in human skeletal muscle. The research, led by Soo-Bin Yang and Hwan Young Lee from Seoul National University College of Medicine, focused on changes in the molecular structure of muscle tissue as people age.
The team analyzed DNA methylation profiles from 103 pectoralis major muscle samples collected during autopsies of South Korean individuals ranging from 18 to 85 years old. "We analyzed DNA methylation profiles from 103 pectoralis major muscle samples from autopsies of South Korean individuals (18–85 years) using the Infinium EPIC array," the authors stated.
Through their analysis, researchers identified 20 specific DNA methylation sites—known as CpGs—that were closely linked to aging. These markers are located within genes related to muscle function, stress response, metabolism, and diseases associated with aging. Using this information, they developed two machine learning models for age prediction: one based on Next-Generation Sequencing (NGS) and another using Single Base Extension (SBE). Both models demonstrated high accuracy, with average errors between 3.8 and 5.5 years.
The newly developed epigenetic clocks performed better than previous models that were designed for other types of tissues. However, when applied to cardiac and uterine muscles, these models showed much lower accuracy. This suggests that molecular age estimation requires approaches tailored to specific tissue types.
In addition to predicting chronological age, the study explored how changes in DNA methylation might influence muscle aging itself. Several CpGs identified were found in regions that regulate gene expression and showed reduced activity in older muscle samples. Some affected genes are known to be linked with sarcopenia—the loss of muscle mass and strength that occurs with aging.
The findings introduce reliable methods for estimating age even when DNA is partially degraded, which may be particularly useful in forensic science. The results also highlight differences in skeletal muscle aging across populations and tissue types and could inform future therapies aimed at slowing age-related decline in muscle health.
