Artificial intelligence-based computational framework for drug-target prioritization and inference of novel repositionable drugs for Alzheimer’s disease

Abstract Background Identifying novel therapeutic targets is crucial for the successful development of drugs. However, cost to experimentally identify huge and only approximately 400 genes are FDA-approved As a result, it inevitable develop powerful computational tools that can potential targets. Fortunately, human protein-protein interaction network (PIN) could be useful resource achieve this objective. Methods In study, we developed deep learning-based framework extracts low-dimensional representations high-dimensional PIN data. Our uses latent features state-of-the-art machine learning techniques infer drug target genes. Results We applied our prioritize putative Alzheimer’s disease successfully identified key may serve as (e.g., DLG4, EGFR, RAC1, SYK, PTK2B, SOCS1). Furthermore, based on these targets, repositionable candidate-compounds tamoxifen, bosutinib, dasatinib). Conclusions tool efficiently new enhance repositioning strategy.