Table 1. Overview of the cohorts for validating DNAmGrimAge2.
| Study | Race2 | N | Female | Death | Age | Follow-up | |
| Samples | Subjects | ||||||
| Training data | |||||||
| FHS training | White | 1833 | 1833 | 54% | 13% | 66.1±9.06 [59,73] | 7.9±1.67 [7.4,9.89] |
| Validation data | |||||||
| FHS test | White | 711 | 711 | 54% | 14% | 66.8±8.62 [61,73] | 7.7±1.77 [7.2,8.78] |
| WHI BA23 | White | 998 | 998 | 100% | 67% | 68.3±6.26 [65,72.77] | 19.1±6.22 [15,23.92] |
| AfricanA | 676 | 676 | 100% | 52% | 63±6.61 [57.9,67.7] | 19.5±6.81 [15.7,24.58] | |
| Hispanic | 433 | 433 | 100% | 43% | 62.2±6.87 [56.5,67.5] | 20.7±5.78 [18.2,24.48] | |
| WHI EMPC | White | 1096 | 1096 | 100% | 48% | 64.3±7.1 [58.9,69.79] | 21±5.95 [18.2,24.96] |
| AfricanA | 558 | 558 | 100% | 45% | 62.5±6.98 [57.7,67.46] | 21±5.67 [18.8,24.77] | |
| Hispanic | 318 | 318 | 100% | 30% | 61.2±6.21 [56.5,65.96] | 22±4.82 [21.9,24.59] | |
| JHS | AfricanA | 1746 | 1746 | 63% | 16% | 56.2±12.32 [46.5,65.35] | 11.7±2.55 [11.2,13.11] |
| InCHIANTI | White | 1430 | 728 | 56% | 37% | 67.4±16.17 [61,78] | 10±4.87 [5.4,14.58] |
| BLSA | White | 572 | 556 | 46% | 32% | 70.9±14.08 [62,82] | 6.1±4.18 [2.1,9.32] |
| LBC21 | White | 692 | 469 | 60% | 94% | 82.3±4.31 [79,86.56] | 8.8±5.2 [4.6,12.57] |
| LBC36 | White | 2796 | 1044 | 50% | 30% | 73.6±3.67 [70.3,76.63] | 9.7±4.09 [6,13.05] |
| NAS | White | 1373 | 732 | 0% | 38% | 74.5±6.99 [69,79] | 10.5±4.71 [6,15] |
| Summary1 | White, African A, Hispanic | 13,399 | 10,065 | 71% | 39% | 67.9±11.33 [61.8,76] | 13±6.9 [7.8,16.91] |
1Summary statistics are based on the nine validation datasets.
2AfricanA denotes African American.
The table summarizes the characteristics of 1833 and 13,399 blood samples used in our training and validation process. The training dataset was based on the 1833 individuals from Framingham Heart Study Offspring Cohort (FHS). The validation datasets consist of 10,065 individuals came from nine independent cohorts: FHS test dataset, Women’s Health Initiatives (WHI) BA23, WHI EMPC, Jackson Heart Study (JHS), InCHIANTI (baseline and the third follow-up), Baltimore Longitudinal Study of Aging (BLSA), Lothian Birth Cohort 1921 (LBC21) and LBC 1936 (LBC36), and Normative Aging Study (NAS). The studies cohorts were used in our validation analysis stratified by racial group within each cohort. Age and follow-up time (from methylation profile to last visit/death in units of years) are presented in the format of mean ±SD [25th, 75th]. Proportion of females are based on individual level.