Improved precision of epigenetic clock estimates across tissues and its implication for biological ageing

Qian Zhang; Costanza L. Vallerga; Rosie M. Walker; Tian Lin; Anjali K. Henders; Grant W. Montgomery; Ji He; Dongsheng Fan; Javed Fowdar; Martin Kennedy; Toni Pitcher; John Pearson; Glenda Halliday; John B. Kwok; Ian Hickie; Simon Lewis; Tim Anderson; Peter A. Silburn; George D. Mellick; Sarah E. Harris; Paul Redmond; Alison D. Murray; David J. Porteous; Christopher S. Haley; Kathryn L. Evans; Andrew M. Mcintosh; Jian Yang; Jacob Gratten; Riccardo E. Marioni; Naomi R. Wray; Ian J. Deary; Allan F. Mcrae; Peter M. Visscher

doi:10.1186/s13073-019-0667-1

Improved precision of epigenetic clock estimates across tissues and its implication for biological ageing

Qian Zhang, Costanza L. Vallerga, Rosie M. Walker, Tian Lin, Anjali K. Henders, Grant W. Montgomery, Ji He, Dongsheng Fan, Javed Fowdar, Martin Kennedy, Toni Pitcher, John Pearson, Glenda Halliday, John B. Kwok, Ian Hickie, Simon Lewis, Tim Anderson, Peter A. Silburn, George D. Mellick, Sarah E. HarrisPaul Redmond, Alison D. Murray, David J. Porteous, Christopher S. Haley, Kathryn L. Evans, Andrew M. Mcintosh, Jian Yang, Jacob Gratten, Riccardo E. Marioni, Naomi R. Wray, Ian J. Deary, Allan F. Mcrae, Peter M. Visscher

Research output: Contribution to journal › Article › peer-review

Abstract

Background
DNA methylation changes with age. Chronological age predictors built from DNA methylation are termed ‘epigenetic clocks’. The deviation of predicted age from the actual age (‘age acceleration residual’, AAR) has been reported to be associated with death. However, it is currently unclear how a better prediction of chronological age affects such association.
Methods
In this study, we build multiple predictors based on training DNA methylation samples selected from 13,661 samples (13,402 from blood and 259 from saliva). We use the Lothian Birth Cohorts of 1921 (LBC1921) and 1936 (LBC1936) to examine whether the association between AAR (from these predictors) and death is affected by (1) improving prediction accuracy of an age predictor as its training sample size increases (from 335 to 12,710) and (2) additionally correcting for confounders (i.e., cellular compositions). In addition, we investigated the performance of our predictor in non-blood tissues.
Results
We found that in principle, a near-perfect age predictor could be developed when the training sample size is sufficiently large. The association between AAR and mortality attenuates as prediction accuracy increases. AAR from our best predictor (based on Elastic Net, https://github.com/qzhang314/DNAm-based-age-predictor) exhibits no association with mortality in both LBC1921 (hazard ratio = 1.08, 95% CI 0.91–1.27) and LBC1936 (hazard ratio = 1.00, 95% CI 0.79–1.28). Predictors based on small sample size are prone to confounding by cellular compositions relative to those from large sample size. We observed comparable performance of our predictor in non-blood tissues with a multi-tissue-based predictor.
Conclusions
This study indicates that the epigenetic clock can be improved by increasing the training sample size and that its association with mortality attenuates with increased prediction of chronological age.

Original language	English
Article number	54
Journal	Genome Medicine
Volume	11
Issue number	1
DOIs	https://doi.org/10.1186/s13073-019-0667-1
Publication status	Published - 23 Aug 2019

Access to Document

10.1186/s13073-019-0667-1Licence: Creative Commons: Attribution (CC-BY)

ZhangEtalBMCGM2019ImprovedPrecisionOfEpigeneticClockFinal published version, 1.25 MBLicence: Creative Commons: Attribution Non-Commercial (CC-BY-NC)

The Disconnected Mind Phase 3
Cox, S. (Principal Investigator) & Deary, I. (Co-investigator)
1/04/16 → 31/03/23
Project: Research
RA2665 Centre for Cognitive Ageing and Cognitive Epidemiology Phase 2.
Wardlaw, J. (Principal Investigator)
1/09/13 → 31/08/19
Project: Research
A genome wide association study of non pathological cognitive ageing
Deary, I. (Principal Investigator), Porteous, D. (Co-investigator) & Tenesa, A. (Co-investigator)
BBSRC
1/09/08 → 31/08/10
Project: Research

Cite this

Zhang, Q., Vallerga, C. L., Walker, R. M., Lin, T., Henders, A. K., Montgomery, G. W., He, J., Fan, D., Fowdar, J., Kennedy, M., Pitcher, T., Pearson, J., Halliday, G., Kwok, J. B., Hickie, I., Lewis, S., Anderson, T., Silburn, P. A., Mellick, G. D., ... Visscher, P. M. (2019). Improved precision of epigenetic clock estimates across tissues and its implication for biological ageing. Genome Medicine, 11(1), Article 54. https://doi.org/10.1186/s13073-019-0667-1

@article{d782c595a6f24a98ab4fc781fa138f1b,

title = "Improved precision of epigenetic clock estimates across tissues and its implication for biological ageing",

abstract = "BackgroundDNA methylation changes with age. Chronological age predictors built from DNA methylation are termed {\textquoteleft}epigenetic clocks{\textquoteright}. The deviation of predicted age from the actual age ({\textquoteleft}age acceleration residual{\textquoteright}, AAR) has been reported to be associated with death. However, it is currently unclear how a better prediction of chronological age affects such association.MethodsIn this study, we build multiple predictors based on training DNA methylation samples selected from 13,661 samples (13,402 from blood and 259 from saliva). We use the Lothian Birth Cohorts of 1921 (LBC1921) and 1936 (LBC1936) to examine whether the association between AAR (from these predictors) and death is affected by (1) improving prediction accuracy of an age predictor as its training sample size increases (from 335 to 12,710) and (2) additionally correcting for confounders (i.e., cellular compositions). In addition, we investigated the performance of our predictor in non-blood tissues.ResultsWe found that in principle, a near-perfect age predictor could be developed when the training sample size is sufficiently large. The association between AAR and mortality attenuates as prediction accuracy increases. AAR from our best predictor (based on Elastic Net, https://github.com/qzhang314/DNAm-based-age-predictor) exhibits no association with mortality in both LBC1921 (hazard ratio = 1.08, 95% CI 0.91–1.27) and LBC1936 (hazard ratio = 1.00, 95% CI 0.79–1.28). Predictors based on small sample size are prone to confounding by cellular compositions relative to those from large sample size. We observed comparable performance of our predictor in non-blood tissues with a multi-tissue-based predictor.ConclusionsThis study indicates that the epigenetic clock can be improved by increasing the training sample size and that its association with mortality attenuates with increased prediction of chronological age.",

author = "Qian Zhang and Vallerga, {Costanza L.} and Walker, {Rosie M.} and Tian Lin and Henders, {Anjali K.} and Montgomery, {Grant W.} and Ji He and Dongsheng Fan and Javed Fowdar and Martin Kennedy and Toni Pitcher and John Pearson and Glenda Halliday and Kwok, {John B.} and Ian Hickie and Simon Lewis and Tim Anderson and Silburn, {Peter A.} and Mellick, {George D.} and Harris, {Sarah E.} and Paul Redmond and Murray, {Alison D.} and Porteous, {David J.} and Haley, {Christopher S.} and Evans, {Kathryn L.} and Mcintosh, {Andrew M.} and Jian Yang and Jacob Gratten and Marioni, {Riccardo E.} and Wray, {Naomi R.} and Deary, {Ian J.} and Mcrae, {Allan F.} and Visscher, {Peter M.}",

year = "2019",

month = aug,

day = "23",

doi = "10.1186/s13073-019-0667-1",

language = "English",

volume = "11",

journal = "Genome Medicine",

issn = "1756-994X",

publisher = "BioMed Central",

number = "1",

}

Zhang, Q, Vallerga, CL, Walker, RM, Lin, T, Henders, AK, Montgomery, GW, He, J, Fan, D, Fowdar, J, Kennedy, M, Pitcher, T, Pearson, J, Halliday, G, Kwok, JB, Hickie, I, Lewis, S, Anderson, T, Silburn, PA, Mellick, GD, Harris, SE, Redmond, P, Murray, AD, Porteous, DJ, Haley, CS, Evans, KL , Mcintosh, AM, Yang, J, Gratten, J, Marioni, RE, Wray, NR, Deary, IJ, Mcrae, AF & Visscher, PM 2019, 'Improved precision of epigenetic clock estimates across tissues and its implication for biological ageing', Genome Medicine, vol. 11, no. 1, 54. https://doi.org/10.1186/s13073-019-0667-1

TY - JOUR

T1 - Improved precision of epigenetic clock estimates across tissues and its implication for biological ageing

AU - Zhang, Qian

AU - Vallerga, Costanza L.

AU - Walker, Rosie M.

AU - Lin, Tian

AU - Henders, Anjali K.

AU - Montgomery, Grant W.

AU - He, Ji

AU - Fan, Dongsheng

AU - Fowdar, Javed

AU - Kennedy, Martin

AU - Pitcher, Toni

AU - Pearson, John

AU - Halliday, Glenda

AU - Kwok, John B.

AU - Hickie, Ian

AU - Lewis, Simon

AU - Anderson, Tim

AU - Silburn, Peter A.

AU - Mellick, George D.

AU - Harris, Sarah E.

AU - Redmond, Paul

AU - Murray, Alison D.

AU - Porteous, David J.

AU - Haley, Christopher S.

AU - Evans, Kathryn L.

AU - Mcintosh, Andrew M.

AU - Yang, Jian

AU - Gratten, Jacob

AU - Marioni, Riccardo E.

AU - Wray, Naomi R.

AU - Deary, Ian J.

AU - Mcrae, Allan F.

AU - Visscher, Peter M.

PY - 2019/8/23

Y1 - 2019/8/23

N2 - BackgroundDNA methylation changes with age. Chronological age predictors built from DNA methylation are termed ‘epigenetic clocks’. The deviation of predicted age from the actual age (‘age acceleration residual’, AAR) has been reported to be associated with death. However, it is currently unclear how a better prediction of chronological age affects such association.MethodsIn this study, we build multiple predictors based on training DNA methylation samples selected from 13,661 samples (13,402 from blood and 259 from saliva). We use the Lothian Birth Cohorts of 1921 (LBC1921) and 1936 (LBC1936) to examine whether the association between AAR (from these predictors) and death is affected by (1) improving prediction accuracy of an age predictor as its training sample size increases (from 335 to 12,710) and (2) additionally correcting for confounders (i.e., cellular compositions). In addition, we investigated the performance of our predictor in non-blood tissues.ResultsWe found that in principle, a near-perfect age predictor could be developed when the training sample size is sufficiently large. The association between AAR and mortality attenuates as prediction accuracy increases. AAR from our best predictor (based on Elastic Net, https://github.com/qzhang314/DNAm-based-age-predictor) exhibits no association with mortality in both LBC1921 (hazard ratio = 1.08, 95% CI 0.91–1.27) and LBC1936 (hazard ratio = 1.00, 95% CI 0.79–1.28). Predictors based on small sample size are prone to confounding by cellular compositions relative to those from large sample size. We observed comparable performance of our predictor in non-blood tissues with a multi-tissue-based predictor.ConclusionsThis study indicates that the epigenetic clock can be improved by increasing the training sample size and that its association with mortality attenuates with increased prediction of chronological age.

AB - BackgroundDNA methylation changes with age. Chronological age predictors built from DNA methylation are termed ‘epigenetic clocks’. The deviation of predicted age from the actual age (‘age acceleration residual’, AAR) has been reported to be associated with death. However, it is currently unclear how a better prediction of chronological age affects such association.MethodsIn this study, we build multiple predictors based on training DNA methylation samples selected from 13,661 samples (13,402 from blood and 259 from saliva). We use the Lothian Birth Cohorts of 1921 (LBC1921) and 1936 (LBC1936) to examine whether the association between AAR (from these predictors) and death is affected by (1) improving prediction accuracy of an age predictor as its training sample size increases (from 335 to 12,710) and (2) additionally correcting for confounders (i.e., cellular compositions). In addition, we investigated the performance of our predictor in non-blood tissues.ResultsWe found that in principle, a near-perfect age predictor could be developed when the training sample size is sufficiently large. The association between AAR and mortality attenuates as prediction accuracy increases. AAR from our best predictor (based on Elastic Net, https://github.com/qzhang314/DNAm-based-age-predictor) exhibits no association with mortality in both LBC1921 (hazard ratio = 1.08, 95% CI 0.91–1.27) and LBC1936 (hazard ratio = 1.00, 95% CI 0.79–1.28). Predictors based on small sample size are prone to confounding by cellular compositions relative to those from large sample size. We observed comparable performance of our predictor in non-blood tissues with a multi-tissue-based predictor.ConclusionsThis study indicates that the epigenetic clock can be improved by increasing the training sample size and that its association with mortality attenuates with increased prediction of chronological age.

U2 - 10.1186/s13073-019-0667-1

DO - 10.1186/s13073-019-0667-1

M3 - Article

SN - 1756-994X

VL - 11

JO - Genome Medicine

JF - Genome Medicine

IS - 1

M1 - 54

ER -

Improved precision of epigenetic clock estimates across tissues and its implication for biological ageing

Abstract

Access to Document

Fingerprint

Projects

The Disconnected Mind Phase 3

RA2665 Centre for Cognitive Ageing and Cognitive Epidemiology Phase 2.

A genome wide association study of non pathological cognitive ageing

Cite this