Rational drug repositioning for coronavirus-associated diseases using directional mapping and side-effect inference

Abstract

SARS-CoV-2, the pathogen of COVID-19, has infected hundreds of millions of people and caused millions of deaths. Looking for valid druggable targets with minimal side effects for the treatment of COVID-19 remains critical. After discovering host genes from multi-scale omics data, we developed an end-to-end network method to investigate drug-host gene(s)-CoV paths and the mechanism of action between the drug and the host factor in a directional network. We also inspected the potential side effect of the candidate drug on several common comorbidities. We established a catalog of host genes associated with three CoVs. Rule-based prioritization yielded 29 FDA-approved drugs via accounting for the effects of drugs on CoVs, comorbidities, and drug-target confidence information. Seven drugs are currently undergoing clinical trials as COVID-19 treatment. This catalog of druggable host genes associated with CoVs and the prioritized repurposed drugs will provide a new sight in therapeutics discovery for severe COVID-19 patients.