Proteomic aging clocks plus deep learning track eye aging and age-related eye disease
In three large cross-national cohorts, machine-learning proteomic aging scores aligned with eye-aging signals and were higher in people with major age-related eye diseases; the abstract reports this as a proposed biomarker, not a proven cause or clinical tool. [1]
Automatic summary · not editorially reviewed
The key findings
The abstract reports three large-scale, cross-national cohorts with over 55,000 transethnic participants, analyzed in discovery and external validation settings. [1]
Premature proteomic aging was identified and validated in people with cataract, diabetic retinopathy, age-related macular degeneration and glaucoma, and proposed as a predictive biomarker beyond chronological age. [1]
Accelerated proteomic aging was linked to neuroretinal degeneration and microvascular rarefaction in two named imaging studies. [1]
The source at a glance
- Study type / format
- Abstract of a multi-cohort observational study using proteomics and deep learning phenotyping
- Population / context
- Over 55,000 transethnic participants in three large-scale cross-national cohorts, plus participants in two named imaging studies.
- Summary based on
- Study abstract
- Original publication
- June 6, 2026
Three cohorts, over 55,000 participants
The abstract describes data from three large-scale, cross-national cohorts with more than 55,000 participants of diverse ethnic backgrounds. The authors combined high-throughput proteomics with deep learning phenotyping, meaning automated image-based characterization, and report results in both discovery and external validation settings. This is an abstract-level description; the full text was not reviewed.
Supporting source [1]
Higher proteomic aging in four age-related eye diseases
Machine-learning proteomic aging scores closely matched eye-aging signals and deep-learning aging phenotypes. The authors report and validate premature proteomic aging in people with major age-related eye diseases: cataract, diabetic retinopathy, age-related macular degeneration and glaucoma. They propose accelerated proteomic aging as a biomarker for predicting these conditions beyond chronological age, with adaptability across sexes and ethnicities. These are associations in observational cohort data, not evidence that proteomic aging causes the diseases.
Supporting source [1]
Imaging links accelerated aging to neural and vascular decline
The authors also describe a streamlined, cost-effective proteomic aging clock that keeps predictive performance while reducing assay complexity. Using tomographic and angiographic imaging processed through deep-learning pipelines, they linked accelerated proteomic aging to neuroretinal degeneration and microvascular rarefaction, meaning loss of small blood vessels, in the Guangzhou Diabetic Eye Study and the High-definition Oculo-Phenomic Evaluation study. The abstract frames the combined approach as a scalable tool for tracking eye health, but does not report clinical use or outcomes.
Supporting source [1]
How to interpret this
This explainer is based only on the abstract; the full text was not reviewed. The abstract does not report study design details, follow-up, effect sizes or how prediction was evaluated. The findings are associations from cohort data, and the biomarker and clock are proposed, not validated clinical products.
Based on the study abstract; the full paper, tables and supplementary data were not reviewed.
Source & supporting passages
Yang S, Xin Z, Li H, Zhu Z, Zhu Z, Chen Y, Zhang Y, Xiong R, Zhou M, He M, Wang W. · europepmc.org · June 6, 2026