Table 3.

The relationship between DNAm age and age (age and age²) using different DNAm clock algorithms

	Testing dataset				Independent test dataset				Blood dataset
	Beta	SE	P	R2	Beta	SE	P	R2	Beta	SE	P	R2
DNAmClockCortical
DNAm age versus age	1.14	3.39 × 10−2	2.86 × 10−108	0.98	1.03	0.17	5.31 × 10−9	0.65	0.58	0.06	5.37 × 10−20	0.78
DNAm age versus age2	−1.64 × 10−3	3.08 × 10−4	1.94 × 10−7	1.57 × 10−3	−2.39 × 10−3	1.08 × 10−3	2.80 × 10−2	1.47 × 10−3	−2.05 × 10−4	5.34 × 10−4	0.70	1.61 × 10−4
DNAmClockMulti
DNAm age versus age	1.08	4.14 × 10−2	3.17 × 10−83	0.93	0.68	0.16	3.51 × 10−5	0.42	0.75	0.06	6.01 × 10−30	0.80
DNAm age versus age2	−3.81 × 10−3	3.75 × 10−4	2.45 × 10−21	0.02	−1.76 × 10−3	1.02 × 10−3	8.50 × 10−2	1.39 × 10−3	−1.15 × 10−3	5.49 × 10−4	0.04	5.66 × 10−4
DNAmClockBlood
DNAm age versus age	0.82	3.41 × 10−2	1.30 × 10−74	0.90	0.64	0.18	3.00 × 10−4	0.26	1.14	0.05	9.50 × 10−111	0.94
DNAm age versus age2	−3.16 × 10−3	3.09 × 10−4	1.81 × 10−21	0.02	−2.08 × 10−3	1.09 × 10−3	5.70 × 10−2	2.30 × 10−3	−2.26 × 10−3	3.90 × 10−4	8.47 × 10−9	1.61 × 10−3
DNAmClockPheno
DNAm age versus age	0.57	6.89 × 10−2	3.19 × 10−15	0.65	−0.35	0.23	1.27 × 10−1	0.10	0.63	0.08	1.86 × 10−13	0.75
DNAm age versus age2	−1.79 × 10−3	6.25 × 10−4	4.47 × 10−3	0.01	3.53 × 10−3	1.42 × 10−3	1.40 × 10−2	0.01	6.22 × 10−4	7.21 × 10−4	0.39	5.41 × 10−5

DNAm age was estimated using our novel cortical clock (DNAmClock_Cortical), Horvath’s multi-tissue clock (DNAmClock_Multi) (Horvath, 2013), Zhang’s elastic net blood clock (DNAmClock_Blood) (Zhang et al., 2019) and Levine’s Pheno Age clock (DNAmClock_Pheno) (Levine et al., 2018) in the ‘testing’ dataset (n = 350 cortical samples), the ‘independent test’ dataset (n = 1221 cortical samples) and the blood dataset (n = 1175 whole blood samples).