Systematic design of drug repurposing-oriented Alzheimer's disease ontology

Alzheimer's Disease (AD) is a neurodegenerative disease characterized by dementia and progressive incapacitation. High prevalence of AD in aged people has created tremendous medical and social burdens. Unfortunately, targeting this devastating disease, FDA-approved medications are only symptomatic relief rather than curative treatment, while de novo drug development is costly, time-consuming, and have encountered several failures in recent clinical trials for novel disease-modifying therapies. Drug repurposing, which develops new uses for existing drugs or recycles late-phase-failed compounds to new indications, might bring a more economical and promising approach. Ontology-based reasoning has been proven effective in drug repurposing research. In this article, we introduce our preliminary efforts on designing an ontology which is specific for AD drug repurposing research, i.e., Drug Repurposing-oriented Alzheimer's Disease Ontology (DROADO). Combining both pre-genomic and post-genomic paradigms for computational drug repurposing, we devise a core knowledge model which comprises the essential elements (drug, gene, pathway, target, etc.) and their possible relations in different levels. To integrate comprehensive and up-to-date biological and pharmaceutical advancement, we adopt a hybrid strategy to populate and enrich the ontology (classes and properties), i.e., direct import from well-curated databases, and automated extraction from high-quality papers, leveraging natural language processing (NLP) approaches and tools. After manual curation and expert review, we conduct an evaluation for its usefulness and community-consensus via Ontokeeper, a semiotic-driven and web-based tool. As an in-depth knowledge base, DROADO would be promising in enabling computational algorithms to realize supervised mining from multi-source and multimodal data and facilitating novel AD drug targets discovery.