Abstract
Developing drugs for treating Alzheimer's disease has been extremely challenging and costly due to limited knowledge of underlying mechanisms and therapeutic targets. To address the challenge in AD drug development, we developed a multi-task deep learning pipeline that learns biological interactions and AD risk genes, then utilizes multi-level evidence on drug efficacy to identify repurposable drug candidates. Using the embedding derived from the model, we ranked drug candidates based on evidence from post-treatment transcriptomic patterns, efficacy in preclinical models, population-based treatment effects, and clinical trials. We mechanistically validated the top-ranked candidates in neuronal cells, identifying drug combinations with efficacy in reducing oxidative stress and safety in maintaining neuronal viability and morphology. Our neuronal response experiments confirmed several biologically efficacious drug combinations. This pipeline showed that harmonizing heterogeneous and complementary data/knowledge, including human interactome, transcriptome patterns, experimental efficacy, and real-world patient data shed light on the drug development of complex diseases.