Abstract
Evaluation of biological age, as opposed to chronological age, is of high relevance for interventions to increase healthy ageing. Highly reproducible age-associated DNA methylation (DNAm) changes can be integrated into algorithms for epigenetic age predictions. These predictors have mostly been trained to correlate with chronological age, but they are also indicative for biological ageing. For example, accelerated epigenetic age of blood is associated with higher risk of all-cause mortality in later life. The perceived age of facial images (face-age) is also associated with all-cause mortality and other ageing-associated traits. In this study, we therefore tested the hypothesis that an epigenetic predictor for biological age might be trained on face-age as a surrogate for biological age rather than on chronological age. Our data demonstrate that facial ageing does not correlate with either the epigenetic clock or blood-based DNAm measures.