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

In the present study, the compounds of the dataset are tested in the The Drug Bank is an online database which contains detailed data about various medications 24 . Today, it is being widely used to facilitate in silico drug target discovery, drug design, drug docking or screening, drug metabolism prediction, drug interaction prediction and for general pharmaceutical education. This database of more than 4900 Drugs is categorized into many different types as Trial stages Drugs, Approved Drugs and Withdrawn Drugs. In this database, more than 45% of drugs are approved for various medication purposes 24 . In this research, we have focused only on the FDA approved drugs for repurposing purpose which are around 2388 with the intention that it will minimize clinical trial in the present situation. These drugs were downloaded in the form of SDFs and after processing, the descriptions generated were taken as the test model for developing the train model which was made on the basis of a database containing the inhibitors of the SARS coronavirus. The developed model has predicted few of the potential drugs.