Proteomic clocks combined with deep learning phenotypes track eye aging and diseases
Protein-based aging signatures combined with eye imaging can predict four major age-related eye diseases more reliably than age alone across diverse populations.
Integrating proteomic aging signatures with AI-derived retinal phenotypes across 55,000+ transethnic participants, this study establishes proteomic aging acceleration as a scalable, ethnicity-adaptable biomarker for predicting four major age-related eye diseases beyond chronological age. A streamlined, cost-effective proteomic clock is proposed that retains predictive performance while reducing assay complexity.
What the study was
- Study design
- Cross-national multi-cohort study (>55,000 transethnic participants); discovery + external validation; proteomic aging clocks + deep learning retinal/optic imaging phenotyping
- Population
- Community-based transethnic populations across three large cohorts (Guangzhou Diabetic Eye Study, High-definition Oculo-Phenomic Evaluation study, and UK Biobank-derived cohort)
- Sample size
- 55000
- Category
- Diagnostics
- Maturity
- Validated
- Journal
- npj Digital Medicine
Why it surfaced
Novel integration of proteomics + AI retinal imaging at scale (>55K participants, cross-national, transethnic validation). Positions proteomic aging acceleration as a cross-disease biomarker framework for major eye conditions. Streamlined clock addresses cost barrier for population deployment.
A plain-language summary of published research — not medical advice. Talk to a clinician about your own care.