DNA Methylation Clocks and Their Predictive Capacity for Aging Phenotypes and Healthspan

Abstract

The number of age predictors based on DNA methylation (DNAm) profile is rising due to their potential in predicting healthspan and application in age-related illnesses, such as neurodegenerative diseases. The cumulative assessment of DNAm levels at age-related CpGs (DNAm clock) may reflect biological aging. Such DNAm clocks have been developed using various training models and could mirror different aspects of disease/aging mechanisms. Hence, evaluating several DNAm clocks together may be the most effective strategy in capturing the complexity of the aging process. However, various confounders may influence the outcome of these age predictors, including genetic and environmental factors, as well as technical differences in the selected DNAm arrays. These factors should be taken into consideration when interpreting DNAm clock predictions. In the current review, we discuss 15 reported DNAm clocks with consideration for their utility in investigating neurodegenerative diseases and suggest research directions towards developing a more optimal measure for biological aging.

Keywords: DNA methylation; age-related disease; biological age; chronological age; neurodegenerative disorders.

Figure 1.

Comparison of chronological vs biological DNAm clocks. Each DNAm clock is developed using a unique training model, including a variable number of CpGs, tissue source of DNA and corresponding age-related measures. While chronological DNAm clocks reflect age-related DNAm changes that are shared between individuals and are expected to reflect the intrinsic aging process, biological DNAm clocks reflect age-related DNAm changes that vary between individuals and are expected to capture associations with specific age-related phenotypes and external drivers that may influence age-related DNAm.

Figure 2.

Gene-overlap among the 14 DNAm clocks. Matrix of pairwise intersections illustrating the relationship between different DNAm clocks based on the percentage of overlapping genes (biological DNAm clocks are indicated in red and chronological clocks in black). Each pairwise intersection, which matches 2 of the variables displayed on the horizontal and vertical axes, was calculated as the percentage of overlap between the genes of the different DNAm clocks and ranges from 0% to 100%. Colour depth indicates the strength of the relationship. Plot generated using the pairwise module of the Intervene Shiny App.