Mapping metabolic aging and disease-associated acceleration using an interpretable NMR-based clock

Ibáñez de Opakua A, Bizkarguenaga M, de Diego A, Conde R, González-Valle B, Gil-Redondo R, Santana T, Seco ML, Cenarro A, Lope V, M Sc. SM, Pelusi M. Sc. S, Habisch H, Haudum C, Verheyen N, Obermayer-Pietsch B, Sorando-Fernandez MP, Arana-Arri E, Meijide S, Salazar AM, Hernandez G, Baldán-Martín M, Suárez-Trujillo F, Audije-Gil J, Arriaga JI, Iriberri M, Moran MÁ, Camino X, Gracia A, Bernardo-Seisdedos G, Sousa A, Oliveira N, Verde I, Zaro Bastanzuri MJ, González Fernández MR, Fernández-Bergés D, Navarrete-Arias L, Llamas-Velasco M, Biccari U, Morales R, Madl T, Unda-Urzaiz M, Carracedo A, Diercks T, Castelló A, Schäfer H, Tasic L, Sanchez-Pernaute R, Buguianessi E, Anstee QM, Embade N, Lu SC, Mera F, Valenti L, Luchinat C, Cibrián D, de la Fuente H, Civeira F, Arenas D, Gisbert JP, Chaparro M, Giskeodegard G, Bathen T, Zuazua Iriondo E, Mato JM, Millet O (2026)

Publication Language: English

Publication Status: Submitted

Publication Type: Unpublished / Preprint

Future Publication Type: Journal article

Publication year: 2026

Open Access Link: https://dcn.nat.fau.eu/wp-content/uploads/MetAge_2026.pdf

Abstract

Biological age captures cumulative physiological decline but remains challenging to quantify across the full adult lifespan with mechanistic and clinical interpretability. Here we develop an interpretable metabolic aging clock based on high-throughput serum nuclear magnetic resonance (NMR) profiling in 29,390 individuals aged 7–106 years across multiple cohorts. 

Using a machine-learning framework that integrates quantified metabolites and NMR-inferred clinical biomarkers, we achieve accurate age prediction (r = 0.88; RMSE  8.7 years) while resolving the multicollinearity that limits interpretability in spectral models. Feature attribution reveals a consistent biological signature dominated by chronic inflammation, kidney function, and energy metabolism, with albumin and erythrocyte sedimentation rate emerging as principal determinants of metabolic aging. Application to eleven pathogenic cohorts (analyzing serum from 3,882 additional patient donors that represent distinctive and most frequent causes of disease), shows that metabolic age acceleration reflects disease-specific distortions rather than uniform aging shifts, where acute inflammatory states, metabolic dysfunction, and organ-specific pathologies exhibit distinct signatures. In a prospective cardiovascular cohort, metabolic age acceleration precedes clinical events, highlighting its potential for early risk stratification. 

Together, these results establish an interpretable, scalable metabolic clock that links systemic biochemical states to aging trajectories and disease risk and provide a framework for precision medicine applications.

Authors with CRIS profile

Involved external institutions

Medizinische Universität Graz Austria (AT) Biocruces Health Research Institute Spain (ES) Hospital Universitario De La Princesa Spain (ES) Universidade Estadual de Campinas (UNICAMP) / University of Campinas Brazil (BR) Instituto de Investigación Sanitaria Aragón (IIS Aragón) Spain (ES) Instituto de Salud Carlos III (Institute of Health Carlos III, ISCIII) Spain (ES) Fondazione IRCCS Ca' Granda - Ospedale Maggiore Policlinico Italy (IT) Deutsches Zentrum für Neurodegenerative Erkrankungen (DZNE) Germany (DE) CIC bioGUNE Spain (ES) Hospital de Basurto Spain (ES) Instituto de Investigación Sanitaria (IIS) Bioaraba Spain (ES) Hospital Universitario de Cruces Spain (ES) Donostia University Hospital / Donostia Unibertsitate Ospitalea Spain (ES) University of Deusto / Universidad de Deusto Spain (ES) University of Beira Interior (UBI) Portugal (PT) Instituto de Investigación Sanitaria Princesa Spain (ES) Hospital Universitario de La Princesa Spain (ES) Bruker BioSpin GmbH Germany (DE) University of Turin / Università degli Studi di Torino (UNITO) Italy (IT) Newcastle University United Kingdom (GB) Cedars-Sinai Medical Center United States (USA) (US) Università degli Studi di Firenze / University of Florence Italy (IT) Northern Research Institute Norway (NO)

How to cite

APA:

Ibáñez de Opakua, A., Bizkarguenaga, M., de Diego, A., Conde, R., González-Valle, B., Gil-Redondo, R.,... Millet, O. (2026). Mapping metabolic aging and disease-associated acceleration using an interpretable NMR-based clock. (Unpublished, Submitted).

MLA:

Ibáñez de Opakua, Alain, et al. Mapping metabolic aging and disease-associated acceleration using an interpretable NMR-based clock. Unpublished, Submitted. 2026.

BibTeX: Download