Skip to main content
. 2022 Jan 24;14(2):623–659. doi: 10.18632/aging.203847

Table 1. Multiple omics make accurate ageing clocks.

Omic N individuals N predictors available N predictors selected r
MetaboAge 2019 - 56 0.21
MS Fatty Acids Lipidomics 952 33 27 0.45
DEXA 1158 28 28 0.66
MS Complex Lipidomics 940 908 130 0.7
NMR Metabolomics 1643 86 81 0.74
UPLC IgG Glycomics 1937 77 50 0.74
GlycanAge 2217 - 3 0.75
Clinomics 1815 13 12 0.8
MS Metabolomics 861 682 181 0.81
DNAme Horvath CpGs 957 333 155 0.93
PEA Proteomics 805 886 203 0.93
Horvath 2013 1065 - 353 0.94
Hannum 2013 1065 - 71 0.95
DNAme Hannum CpGs 1033 62 50 0.96
Mega Omics 796 2471 214 0.97

Indicating for each omics assay: N Individuals: the number of individuals in the ORCADES cohort that passes quality control, N Predictors Available: the number of predictors passing assay-level quality control and therefore available for selection for inclusion in the standard model, N Predictors Selected: the number of predictors selected for inclusion in the standard model, r: Pearson correlation of omics clock age (OCA) and chronAge. DEXA, Dual X-ray absorptiometry; DNAme, DNA methylation; CpG, cytosine nucleotide followed by guanine (5’ to 3’ direction); MS, mass spectrometry; NMR, nuclear magnetic resonance; PEA, proximity extension assay; UPLC, ultra-performance liquid chromatography; IgG, Immunoglobulin G. Within each omic category, subject mean age at baseline was 53-56 (SD~15) with an age range across clocks of 16-100, whilst the proportion female ranged from 55-61% (Supplementary Table 1).