Repurposing Therapeutics for COVID-19: Rapid Prediction of Commercially available drugs through Machine Learning and Docking

Sovesh Mahapatra; Prathul Nath; Manisha Chatterjee; Neeladrisingha Das; Deepjyoti Kalita; Partha Roy; Soumitra Satapathi

The copyright holder for this preprint (which was not peer-reviewed) is . https://doi.org/10.1101/2020.04.05.20054254 doi: medRxiv preprint data and thereby make intelligent decisions based on independent datasets 21 . ML can accurately predict drug-target interactions as an enormous amount of complex information by studying hydrophobic interactions, ionic interactions, hydrogen bonding, van der Waals forces, etc. between molecules. Bioactivity datasets which are available from the numerous high throughput screens deliver useful means for machine learning classifiers as they contain binary information (active/inactive) as well as numerical values to classify different compounds under consideration 22, 23 . Such a huge number of datasets available on biological activities of molecules, derived from high throughput screens now allows to create predictive computational models.