Skip to main content
Contact Info
Anshu Bhardwaj
BioiInformatics Centre, CSIR-Institute of Microbial Technology, Chandigarh, India.

SAVIOR - SARS-CoV-2 Inhibitor Open Repository 

Emergence of new coronavirus identified as SARS-CoV-2 has brought the world to a standstill. With no  therapeutic solutions, continuous efforts have been made to find a cure for COVID-19 infection. As  researchers are exploring several methods to identify potential lead candidates, a lot of inhibitors have been  reported against SARS-CoV-2 leading to a continuously growing chemical space which remains largely  unmapped and unanalysed. The purpose of our research was to create a knowledgebase for SARS-CoV-2  inhibitors and map its chemical space to understand its diversity, physicochemical properties and scope for  further development. Towards this, we have crowdsourced over 3000 research papers from a variety of  sources like PubMed, BioRxiv, medRxiv, ChemRxiv and Preprints servers to comb for chemical structure  and associated biological activity data. Data on SARS-CoV-2 inhibitors is also available from several  ongoing open challenges like PostERA, high-throughput screen datasets available from ChEMBL and  NCATS and also CAS contributed an antiviral library of over 49,000 compounds for COVID research.  However, there is no single resource to comprehend or evaluate these independent and fast-paced  contributions to the SARS-CoV-2 chemical space. As of now SAVIOR includes compounds undergoing  clinical trials or have reported in-vivo, in-vitro activity or predicted via in-silico methods. It also includes  over 250 compounds on eleven viral and six host targets. A rigorous data analysis was done with curated  SAVIOR for prioritizing drug-like compounds using a combined evidence approach including structure  similarity of SAVIOR with libraries like COCONUT, PostEra, Pathogen Box, CAS & Prestwick (total library size > 450,000). Comparative analysis was performed on the basis of physicochemical properties  with PostEra compound and CAS antivirals. Several filters including PAINS are used to identify drug-like  compounds from a library of 750 compounds curated so far. To extend the chemical space, similar  compounds (TC ≥0.85) to the six compounds identified in this study are extracted from ChEMBL and  ZINC libraries. SAVIOR is an OPEN data platform for continuous curation and analyses and is built on  FAIR data principles.

The platform may be accessed from