Background Statistical models (known as epigenetic clocks) that use an individual’s DNA methylation to predict their age have recently been developed, with 96% correlation found between predicted and chronological age. Departures of predictions from actual age, called age acceleration residuals, can be used as a measure of biological ageing. We aimed to investigate whether age acceleration was associated with development during childhood and adolescence using repeated measures data from a large birth cohort in the UK.
Methods We obtained DNA methylation at three clinics (birth, age 7 and 17) in 1018 children from the Avon Longitudinal Study of Parents and Children (ALSPAC). We used the epigenetic clock algorithm to obtain DNA methylation age for each child at each clinic. Age acceleration residuals were obtained for each sample as a measure of the difference between predicted and actual age. We then investigated whether age acceleration was associated with repeated measures of height, weight, BMI, bone mineral density, bone mass, fat mass or lean mass. Multilevel models were used to account for both between- and within-individual correlation in the birth cohort, and spline terms were used to account for nonlinear trajectories over time.
Results In ALSPAC, the epigenetic clock predicted actual age with a correlation of 96% and an average absolute difference between predicted and actual age of 2.9 years. Positive age acceleration at birth was associated with higher average height (0.65 cm per year of age acceleration, 95% CI 0.02 cm–1.29 cm; p = 0.044) and fat mass (1.39 kg per year of age acceleration, 95% CI 0.54–2.34 kg; p = 0.002) during childhood and adolescence (i.e., from age 0–17).
Discussion The epigenetic clock represents a useful application of DNA methylation data, with a correlation of 96% between chronological and predicted age. We have found evidence that being ahead of one’s epigenetic clock at birth is associated with increased height and fat mass. This demonstrates the potential of using age acceleration as a measure of development in future research. Our analysis was limited in having data available until 17 rather than into adulthood, but our strengths include repeated measurements of both developmental responses and DNA methylation.