DNA methylation aging clocks: challenges and recommendations

Abstract

Epigenetic clocks comprise a set of CpG sites whose DNA methylation levels measure subject age. These clocks are acknowledged as a highly accurate molecular correlate of chronological age in humans and other vertebrates. Also, extensive research is aimed at their potential to quantify biological aging rates and test longevity or rejuvenating interventions. Here, we discuss key challenges to understand clock mechanisms and biomarker utility. This requires dissecting the drivers and regulators of age-related changes in single-cell, tissue- and disease-specific models, as well as exploring other epigenomic marks, longitudinal and diverse population studies, and non-human models. We also highlight important ethical issues in forensic age determination and predicting the trajectory of biological aging in an individual.

Fig. 1

a Chronological age estimation error. With increasing training sample size, improved measurement of chronological age is expected, even using current array data (adapted from Zhang et al. [46]). y-axis: root mean square error (RMSE) of the predicted age. b DNA methylation clocks contain both chronological and biological information. The relative proportions of each will depend on the CpG probes employed in the construction of the clock. Therefore, there are multiple clocks that can be deconvoluted from aging-related epigenetic changes. Moving forward, more precise chronological (forensic age clock) and biological clocks, specific for particular diseases, informative of health or disease state need to be defined and separated. c Epigenetic age trajectory. Epigenetic age is not linear over the life course. Chronological age in years (x-axis) and epigenetic age in years (y-axis)

Fig. 2

All clock probes are strongly biased to reside within active functional loci. This is due to their construction from promoter-focused arrays. Overlap of CpGs from four DNA methylation clocks with the six Core Encode Combined Chromatin Segmentation tracks [130] from ENCODE Analysis Data at UCSC. a Horvath clock [24]. b Hannum et al. clock [23]. c PhenoAge clock [43]. d epiTOC clock [31]. Location is assessed for overlap with the seven functional categories: PF (promoter flanking—light red), TSS (transcription start site and promoter region—red), CTCF (blue), WE (weak enhancer—yellow), E (enhancer—gold), T (transcribed region—green), and R (repressed—grey), from any of the six Core Encode cell types (Gm12878, H1hesc, Helas3, Hepg2, Huvec, K562). This percentage overlap is shown on the y-axis and is compared with the percentage overlap for all ~28 × 10⁶ CpGs in the human genome on the x-axis. Calculated via bedtools [131]. The size of the circle is proportional to the entire genome space for each functional category (~10^{(genome size proportion)}). e-h Direct overlap comparison for four DNA methylation clocks (Horvath clock, Hannum et al. clock, PhenoAge clock, epiTOC clock) as well as Illumina array CpGs (27k, 450k, EPIC) and all genomic CpGs (far right bar) with: e the Combined Segmentation track for blood-derived tissue (GM12878) [130]. Functional segments are delineated as PF (promoter flanking), TSS (transcription start site and promoter region), CTCF, WE (weak enhancer), E (enhancer), T (transcribed region), and R (repressed). NC, not covered CpGs in this Combined Segmentation overlap; f Gencode [132] Exon and Transcripts; g UCSC [133]-defined CpG islands and shore regions (+/−2 kb); h Major repeat classes (UCSC RepeatMasker [133]), including DNA repeat elements (DNA_repeats), long interspersed nuclear elements (LINE), low complexity repeats and other rare repeat classes, long terminal repeat elements (LTR), simple repeats (microsatellites aka short tandem repeats), and short interspersed nuclear elements (SINE), of which ~63% are Alu elements

Fig. 3

Single-cell analysis. Distinct cell variation in aging epigenetic clock changes may exist that would not be apparent in bulk comparison. Black and white squares represent methylated and unmethylated loci, respectively. Each row represents a single cell’s epigenome (represented as haploid for simplicity) with increased variability present in individual 2