September 10th, 2026
Aging clocks built using machine learning techniques applied to biological data are colloquially said to measure biological age, or at least aspire to measure biological age. Biological age is nebulous and debated at the detail level, but a rigorous definition is the increase over time in risk of mortality from intrinsic causes. A greater mortality risk implies a greater biological age, regardless of chronological age. Regarding clocks, it would be more accurate to say that they may measure some reflection of biological age, or factors that correlate with biological age, and it is actually far from clear that any given clock does this in a useful way. The lengthy review and discussion noted here is a good one, and digs into many of the subtle issues in how aging clocks are presently presented.
Biological age (BA) has been proposed as a complementary construct to chronological age (CA) for quantifying interindividual heterogeneity in aging trajectories. Biological aging clocks (BACs) integrate molecular, clinical and multi-omics biomarkers to estimate aging-related phenotypes beyond CA. This narrative review critically examines the biological foundations, statistical methodologies, interpretative challenges and translational applications of BACs. We discuss the mechanistic basis of BACs development within the frameworks of the hallmarks and domains of aging, emphasizing the roles of age-related methylome remodeling, immunosenescence, and chronic low-grade inflammation.
