Multiscale Interactome Analysis Prioritizes Candidate Herbs and Compounds for Diabetic Retinopathy

Diabetic retinopathy (DR), a leading cause of preventable vision loss among working-age adults, arises from chronic hyperglycemic injury to the retinal microvasculature; current interventions such as anti-vascular endothelial growth factor (VEGF) injection are largely reactive, and mechanism-anchored herbal options remain undefined. We constructed a multiscale interactome and applied a biased random-walk diffusion model to prioritize herbal candidates, candidate compounds, and mechanisms for DR. Herb-ingredient records were drawn from OASIS; ingredient targets from DrugBank, TTD, and STITCH; and disease genes from DisGeNET. For each herb and compound, we computed diffusion profiles, correlated them with the DR profile, and assessed target-overlap significance by hypergeometric test. Eleven herbs combining high correlation (top 5%) with significant overlap (p < 0.05) were retained, led by Pini Koraiensis Semen (0.0263), Benincasae Exocarpium (0.0183), and Veratri Rhizoma Et Radix (0.0162); none had been previously investigated for DR. CASP3 emerged as a target shared by all eleven herbs, and enrichment analysis highlighted the advanced glycation end product-receptor for AGE (AGE-RAGE), TNF, IL-17, and NF-κB pathways alongside apoptosis and reactive-oxygen-species responses. Compound-level analysis recovered evidence-supported quercetin, resveratrol, oleic acid, luteolin, and diosgenin and nominated tectorigenin and cucurbitacins as novel candidates. This framework recapitulates known DR biology and elevates testable multi-target leads that warrant experimental validation for diabetic retinopathy.