ProtFI, an efficient frailty-trained proteomics-based biomarker of aging, robustly predicts age-related decline

None, None; None, None; None, None; None, None; None, None; None, None; None, None; None, None; None, None; None, None

ProtFI, an efficient frailty-trained proteomics-based biomarker of aging, robustly predicts age-related decline

Journal Article (2026)

Author(s)

Swier Garst (TU Delft - Pattern Recognition and Bioinformatics, Leiden University Medical Center)

Lieke Kuiper (Rijksinstituut voor Volksgezondheid en Milieu (RIVM), TU Delft - Pattern Recognition and Bioinformatics, Erasmus MC)

Erik van den Akker (TU Delft - Pattern Recognition and Bioinformatics, Leiden University Medical Center)

Niels van den Berg (Leiden University Medical Center)

Mohsen Ghanbari (Erasmus MC)

Simon Mooijaart (Leiden University Medical Center)

Marian Beekman (Leiden University Medical Center)

Marcel Reinders (TU Delft - Pattern Recognition and Bioinformatics)

P. Eline Slagboom (Leiden University Medical Center)

Joyce van Meurs (Erasmus MC)

Research Group

Pattern Recognition and Bioinformatics

Deep learning Prediction Biomarker Metabolomics Aging Proteomics Frailty CP: computational biology CP: systems biology ElasticNet

DOI related publication

https://doi.org/10.1016/j.crmeth.2026.101405 Final published version

To reference this document use

https://resolver.tudelft.nl/uuid:e9cf4fe9-85d9-4c84-9047-d91f83373bac

More Info

expand_more

Publication Year

2026

Language

English

Research Group

Pattern Recognition and Bioinformatics

Journal title

Cell Reports Methods

Article number

101405

Downloads counter

9

Abstract

Many molecular aging biomarkers have been developed to capture heterogeneity in individual aging rates. Yet, systematic comparison of the modeling choices underlying these biomarkers has been limited. In this study, we trained aging biomarkers on the Rockwood frailty index (FI) and all-cause mortality using UK Biobank Olink proteomics and metabolomics (¹H-NMR) data (n = 40,696). We systematically established the impact of model choice, target outcome, and molecular data source on several age-related outcomes. From this, we developed two aging biomarkers, ProteinFrailty (ProtFI) and ProteinMortality (ProtMort), which are both ElasticNet models that use a minimal set of proteins to predict FI and mortality, respectively. In particular, ProtFI outperformed established aging biomarkers in relation to diverse outcomes, including incident cardiovascular disease, handgrip strength, and self-rated health, both in internal validation and two Dutch external cohorts (n = 995, n = 500). Our findings show that an efficient frailty-trained proteomic biomarker robustly predicts age-related decline.