Proteomic clocks combined with deep learning phenotypes track eye aging and diseases
Shaopeng Yang1, Zhuoyao Xin2,3, Huangdong Li1
1State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Guangdong Provincial Key Laboratory of Ophthalmology and Visual Science, Guangdong Provincial Clinical Research Center for Ocular Diseases, Guangdong Basic Research Center of Excellence (GBRCE) for Major Blinding Eye Diseases Prevention and Treatment, Sun Yat-Sen University, Guangzhou, China, Guangzhou, China.
NPJ Digital Medicine
|June 6, 2026
Summary
Proteomics and AI accurately track eye aging and predict age-related eye diseases. A new, cost-effective proteomic clock shows promise for monitoring eye health across diverse populations.
Area of Science:
- Ophthalmology
- Gerontology
- Biomarker Discovery
Background:
- Eye aging is complex, involving molecular and structural changes.
- Proteomics offers insights into aging mechanisms but is underutilized.
- Current methods for tracking eye aging and disease risk have limitations.
Purpose of the Study:
- To leverage high-throughput proteomics and deep learning (DL) for eye aging and disease characterization.
- To validate proteomic aging as a biomarker for age-related eye diseases (AREDs).
- To develop a cost-effective proteomic aging clock and link it to neurovascular decline.
Main Methods:
- Analysis of proteomic data from over 55,000 participants across three cohorts.
- Application of machine learning and DL for phenotyping and biomarker development.
- Integration of advanced imaging (tomography, angiography) with proteomic data.
Main Results:
- Proteomic aging closely correlates with established eye aging phenotypes.
- Premature proteomic aging was identified in individuals with cataract, diabetic retinopathy, age-related macular degeneration, and glaucoma.
- A streamlined proteomic aging clock demonstrated predictive performance and linked accelerated aging to neuroretinal degeneration and microvascular rarefaction.
Conclusions:
- Proteomic aging, enhanced by AI, is a scalable tool for tracking eye health and disease.
- Accelerated proteomic aging serves as a robust biomarker for predicting AREDs beyond chronological age.
- Findings reveal coupled neuro-vascular decline in eye aging and shared aging pathways across ocular pathologies.
