TABLE 1.

Summary of epigenetic clocks used in the study

Clock	Type	Tissue	Outcome	Publication	Notes
Horvath	Intrinsic	Multiple	Chronological age	Horvath (2013)	Inaccessible tissues primarily from tissue‐adjacent normal samples in The Cancer Genome Atlas (see publication)
IEAA	Intrinsic	Multiple	Chronological age	Quach et al. (2017)	Uses same CpGs as Horvath clock, but reweighted as described in Quach et al. to minimize influence of cell composition
Hannum	Extrinsic	Whole blood	Chronological age	Hannum et al. (2013)	Highly correlated with aging‐related changes in blood cell composition
EEAA	Extrinsic	Whole blood	Chronological age	Quach et al. (2017)	Uses same CpGs as Hannum clock, but reweighted as described in Quach et al. to maximize influence of cell composition
SkinAndBloodClock	Intrinsic	Whole blood, fibroblasts	Chronological age	Horvath et al. (2018)	Created to address poor age prediction in Horvath clock in skin and whole blood
PhenoAge	Extrinsic	Whole blood	Time to death	Levine et al. (2018)	PhenoAge is measure of mortality risk derived from National Health and Nutrition Examination Survey using the following markers: albumin, creatinine, serum glucose, log C‐reactive protein, lymphocyte percent, mean red cell volume, red cell distribution width, alkaline phosphatase, white blood cell count, and age (see publication for details)
GrimAge	Extrinsic	Whole blood	Time to death	Lu, Quach, et al. (2019)	Methylation is used to predict eight surrogate biomarkers: Adrenomedullin (ADM), Beta‐2‐Microglobulin (B2M), Cystatin C, Growth Differentiation Factor 15 (GDF15), Leptin, Serpin Family E Member 1 (SERPINE/PAI1), TIMP Metalloproteinase Inhibitor 1 (TIMP1), smoking pack‐years (PACKYRS). The predicted values of those biomarkers are used to predict mortality (see publication for details)

Abbreviations: EEAA, extrinsic epigenetic age acceleration; IEAA, intrinsic epigenetic age acceleration.