Topic: COVID19 drug repurposing using knowledge graphs
Lead: Halil Kilicoglu
Time: 11 am-noon (CT), Wednesday, 2021-04-07
Zoom: Link
Related Materials: [Box folder]
Abstract:
In this seminar, I will discuss our recent work in discovering candidate drugs to repurpose for COVID19 using knowledge mined from the scientific literature and knowledge graph completion methods. Our approach relies on semantic triples extracted from the literature using the SemRep NLP tool (via SemMedDB). We identified an informative and accurate subset of semantic triples using filtering rules and an accuracy classifier developed on a BERT variant. We used this subset to construct a knowledge graph, and applied five state-of-the-art, neural knowledge graph completion algorithms to predict drug repurposing candidates. The models were trained and assessed using a time slicing approach and the predicted drugs were compared with a list of drugs reported in the literature and evaluated in clinical trials. These models were complemented by a discovery pattern-based approach, which allowed us to generate plausible mechanistic explanations for the candidate drugs.