iLDA-SGCN is a computational framework that predicts associations between long non-coding RNAs and age-related diseases with high accuracy, identifying 33 candidate lncRNAs across eight disease categories. This work systematically maps regulatory pathways that contribute to aging-related disease progression, establishing a foundation for targeted mechanistic investigation.
Key Points
- Model achieved 0.960 AUC on benchmark datasets, outperforming five competitive methods
- Identified 33 candidate lncRNAs across eight age-related diseases requiring validation
- Integrates semantic similarity and network topology for robust association prediction
Longevity Analysis
Age-related diseases emerge from dysregulated cellular and molecular processes that accumulate over time. This computational approach reveals which non-coding RNA regulators are involved in aging phenotypes, offering a mechanism-first entry point for intervention. By systematically mapping these associations rather than studying diseases in isolation, the framework supports a more integrated understanding of how aging manifests across multiple physiological domains. The candidate lncRNAs—particularly HOTAIR, MALAT1, and PVT1—represent testable molecular targets that could be addressed through regenerative support, stress-response modulation, or targeted detoxification strategies once their specific roles in disease pathogenesis are experimentally confirmed.
Original published by Wiley Aging Cell, by Yu Guo, Shizheng Qiu, Zhishuai Zhang, Jirui Guo, Haozheng Liang, Huanyu You, Fengjuan Lu, Yanwei Xu, Yang Hu .