Category: T2 - General Drug Discovery, Including COVID PS ID : DDT2-03

Building SARS-CoV-2 Inhibitor Knowledgebase – SAVIOR

Create an open benchmark repository of SARS-CoV-2 inhibitors. So far we have curated a list of over 450 proposed SARS-CoV-2 inhibitors from several resources like Pubmed, BioRxiv, ChemRxiv, preprint servers, open challenges and other resources. This data may aid in understanding the chemical space being pursued for SARS-CoV-2. Given the rate at which the publications and other efforts are leading to ever increasing data on new chemical entities for SARS-CoV-2, it is imperative that we enrich and analyze SAVIOR. For the same, the participants are expected to: i) Perform automated search using standard MeSH entry terms across several resources listed above to curate compounds. ii) Apply NLP methods like named entity recognition to recognize chemical and biological entities and extract meaningful information for further curation. Tool like Stanford Open Information extractor can be used to to mine abstracts, full-length publications and text in databases, iii) The library once curated will be programmatically assessed for drug-likeness properties using several filters like Lipinski's rule, Veber rule, Ghose filter, etc and clustering will be performed using Jarvis-Patrick clustering algorithm and iv) the compounds predicted or identified as anti-SARS-CoV-2, (during this hackathon) may be compared against the listed compounds at tanimoto coefficient >= 0.6 using FP2 and MACCS fingerprint

Login to Download Input Form