Deep biomarkers of human aging: Application of deep neural networks to biomarker development

Evgeny Putin, Polina Mamoshina, Alexander Aliper, Mikhail Korzinkin, Alexey Moskalev, Alexey Kolosov, Alexander Ostrovskiy, Charles Cantor, Jan Vijg, Alex Zhavoronkov

Research output: Contribution to journalArticle

69 Citations (Scopus)

Abstract

One of the major impediments in human aging research is the absence of a comprehensive and actionable set of biomarkers that may be targeted and measured to track the effectiveness of therapeutic interventions. In this study, we designed a modular ensemble of 21 deep neural networks (DNNs) of varying depth, structure and optimization to predict human chronological age using a basic blood test. To train the DNNs, we used over 60,000 samples from common blood biochemistry and cell count tests from routine health exams performed by a single laboratory and linked to chronological age and sex. The best performing DNN in the ensemble demonstrated 81.5 % epsilon-accuracy r = 0.90 with R2 = 0.80 and MAE = 6.07 years in predicting chronological age within a 10 year frame, while the entire ensemble achieved 83.5% epsilon-accuracy r = 0.91 with R2 = 0.82 and MAE = 5.55 years. The ensemble also identified the 5 most important markers for predicting human chronological age: albumin, glucose, alkaline phosphatase, urea and erythrocytes. To allow for public testing and evaluate real-life performance of the predictor, we developed an online system available at http://www.aging.ai. The ensemble approach may facilitate integration of multi-modal data linked to chronological age and sex that may lead to simple, minimally invasive, and affordable methods of tracking integrated biomarkers of aging in humans and performing cross-species feature importance analysis.

Original languageEnglish (US)
Pages (from-to)1021-1033
Number of pages13
JournalAging
Volume8
Issue number5
StatePublished - 2016

Fingerprint

Biomarkers
Online Systems
Blood Cell Count
Hematologic Tests
Biochemistry
Alkaline Phosphatase
Urea
Albumins
Erythrocytes
Glucose
Health
Research
Therapeutics

Keywords

  • Aging biomarkers
  • Biomarker development
  • Deep learning
  • Deep neural networks
  • Human aging
  • Machine learning

ASJC Scopus subject areas

  • Aging
  • Cell Biology

Cite this

Putin, E., Mamoshina, P., Aliper, A., Korzinkin, M., Moskalev, A., Kolosov, A., ... Zhavoronkov, A. (2016). Deep biomarkers of human aging: Application of deep neural networks to biomarker development. Aging, 8(5), 1021-1033.

Deep biomarkers of human aging : Application of deep neural networks to biomarker development. / Putin, Evgeny; Mamoshina, Polina; Aliper, Alexander; Korzinkin, Mikhail; Moskalev, Alexey; Kolosov, Alexey; Ostrovskiy, Alexander; Cantor, Charles; Vijg, Jan; Zhavoronkov, Alex.

In: Aging, Vol. 8, No. 5, 2016, p. 1021-1033.

Research output: Contribution to journalArticle

Putin, E, Mamoshina, P, Aliper, A, Korzinkin, M, Moskalev, A, Kolosov, A, Ostrovskiy, A, Cantor, C, Vijg, J & Zhavoronkov, A 2016, 'Deep biomarkers of human aging: Application of deep neural networks to biomarker development', Aging, vol. 8, no. 5, pp. 1021-1033.
Putin E, Mamoshina P, Aliper A, Korzinkin M, Moskalev A, Kolosov A et al. Deep biomarkers of human aging: Application of deep neural networks to biomarker development. Aging. 2016;8(5):1021-1033.
Putin, Evgeny ; Mamoshina, Polina ; Aliper, Alexander ; Korzinkin, Mikhail ; Moskalev, Alexey ; Kolosov, Alexey ; Ostrovskiy, Alexander ; Cantor, Charles ; Vijg, Jan ; Zhavoronkov, Alex. / Deep biomarkers of human aging : Application of deep neural networks to biomarker development. In: Aging. 2016 ; Vol. 8, No. 5. pp. 1021-1033.
@article{9035488896714e64aad1ddd408c94106,
title = "Deep biomarkers of human aging: Application of deep neural networks to biomarker development",
abstract = "One of the major impediments in human aging research is the absence of a comprehensive and actionable set of biomarkers that may be targeted and measured to track the effectiveness of therapeutic interventions. In this study, we designed a modular ensemble of 21 deep neural networks (DNNs) of varying depth, structure and optimization to predict human chronological age using a basic blood test. To train the DNNs, we used over 60,000 samples from common blood biochemistry and cell count tests from routine health exams performed by a single laboratory and linked to chronological age and sex. The best performing DNN in the ensemble demonstrated 81.5 {\%} epsilon-accuracy r = 0.90 with R2 = 0.80 and MAE = 6.07 years in predicting chronological age within a 10 year frame, while the entire ensemble achieved 83.5{\%} epsilon-accuracy r = 0.91 with R2 = 0.82 and MAE = 5.55 years. The ensemble also identified the 5 most important markers for predicting human chronological age: albumin, glucose, alkaline phosphatase, urea and erythrocytes. To allow for public testing and evaluate real-life performance of the predictor, we developed an online system available at http://www.aging.ai. The ensemble approach may facilitate integration of multi-modal data linked to chronological age and sex that may lead to simple, minimally invasive, and affordable methods of tracking integrated biomarkers of aging in humans and performing cross-species feature importance analysis.",
keywords = "Aging biomarkers, Biomarker development, Deep learning, Deep neural networks, Human aging, Machine learning",
author = "Evgeny Putin and Polina Mamoshina and Alexander Aliper and Mikhail Korzinkin and Alexey Moskalev and Alexey Kolosov and Alexander Ostrovskiy and Charles Cantor and Jan Vijg and Alex Zhavoronkov",
year = "2016",
language = "English (US)",
volume = "8",
pages = "1021--1033",
journal = "Aging",
issn = "1945-4589",
publisher = "US Administration on Aging",
number = "5",

}

TY - JOUR

T1 - Deep biomarkers of human aging

T2 - Application of deep neural networks to biomarker development

AU - Putin, Evgeny

AU - Mamoshina, Polina

AU - Aliper, Alexander

AU - Korzinkin, Mikhail

AU - Moskalev, Alexey

AU - Kolosov, Alexey

AU - Ostrovskiy, Alexander

AU - Cantor, Charles

AU - Vijg, Jan

AU - Zhavoronkov, Alex

PY - 2016

Y1 - 2016

N2 - One of the major impediments in human aging research is the absence of a comprehensive and actionable set of biomarkers that may be targeted and measured to track the effectiveness of therapeutic interventions. In this study, we designed a modular ensemble of 21 deep neural networks (DNNs) of varying depth, structure and optimization to predict human chronological age using a basic blood test. To train the DNNs, we used over 60,000 samples from common blood biochemistry and cell count tests from routine health exams performed by a single laboratory and linked to chronological age and sex. The best performing DNN in the ensemble demonstrated 81.5 % epsilon-accuracy r = 0.90 with R2 = 0.80 and MAE = 6.07 years in predicting chronological age within a 10 year frame, while the entire ensemble achieved 83.5% epsilon-accuracy r = 0.91 with R2 = 0.82 and MAE = 5.55 years. The ensemble also identified the 5 most important markers for predicting human chronological age: albumin, glucose, alkaline phosphatase, urea and erythrocytes. To allow for public testing and evaluate real-life performance of the predictor, we developed an online system available at http://www.aging.ai. The ensemble approach may facilitate integration of multi-modal data linked to chronological age and sex that may lead to simple, minimally invasive, and affordable methods of tracking integrated biomarkers of aging in humans and performing cross-species feature importance analysis.

AB - One of the major impediments in human aging research is the absence of a comprehensive and actionable set of biomarkers that may be targeted and measured to track the effectiveness of therapeutic interventions. In this study, we designed a modular ensemble of 21 deep neural networks (DNNs) of varying depth, structure and optimization to predict human chronological age using a basic blood test. To train the DNNs, we used over 60,000 samples from common blood biochemistry and cell count tests from routine health exams performed by a single laboratory and linked to chronological age and sex. The best performing DNN in the ensemble demonstrated 81.5 % epsilon-accuracy r = 0.90 with R2 = 0.80 and MAE = 6.07 years in predicting chronological age within a 10 year frame, while the entire ensemble achieved 83.5% epsilon-accuracy r = 0.91 with R2 = 0.82 and MAE = 5.55 years. The ensemble also identified the 5 most important markers for predicting human chronological age: albumin, glucose, alkaline phosphatase, urea and erythrocytes. To allow for public testing and evaluate real-life performance of the predictor, we developed an online system available at http://www.aging.ai. The ensemble approach may facilitate integration of multi-modal data linked to chronological age and sex that may lead to simple, minimally invasive, and affordable methods of tracking integrated biomarkers of aging in humans and performing cross-species feature importance analysis.

KW - Aging biomarkers

KW - Biomarker development

KW - Deep learning

KW - Deep neural networks

KW - Human aging

KW - Machine learning

UR - http://www.scopus.com/inward/record.url?scp=84976407069&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84976407069&partnerID=8YFLogxK

M3 - Article

C2 - 27191382

AN - SCOPUS:84976407069

VL - 8

SP - 1021

EP - 1033

JO - Aging

JF - Aging

SN - 1945-4589

IS - 5

ER -