Integrating generative AI and genetic causal inference identifies metformin as a repurposing candidate for normal tension glaucoma.

The normal tension glaucoma (NTG) has limited drug options since current antiglaucoma medications are mostly designed to decrease intraocular pressure. The emerging generative artificial intelligence (GAI) may provide an unprecedented approach for the drug repurposing research of NTG. We interacted with three GAI models to offer 20 drug repurposing candidates (DRCs) for NTG. By using GAI and DrugBank database, the targets for the selected DRCs were identified. Then, the ChEMBL database was utilized to find the target-associated genes. The relevant instrumental variables (IVs) mapped to these genes were identified with the GTEX dataset. The mediation exposures (e.g., HbA1c for metformin) for the identified drugs were introduced to the single SNP Mendelian randomization to filter the IVs with significant causal influence on the mediation traits. The filtered IVs were utilized to measure the DRCs' causal effect on NTG. Applying the target-based MR, we found that metformin may exert causal influence on NTG through targets GLP-1 and gluconeogenic enzymes. A literature-based evidence assessment supported these findings, with the most identified studies indicating a protective role of metformin in glaucoma-related outcomes. Furthermore, disease-centric interaction analysis using the DisGeNET database revealed that 'disorder of optic nerve' ranked among the top categories associated with metformin's targets, providing additional biological plausibility for its relevance to NTG. Our results not only offered novel evidence to support the metformin's repurposing in NTG but also proposed a paradigm consisting of GAI and genetic tools, which could serve as a pipeline for drug repurposing of other diseases.