COVID-19 Drug Discovery and Comparison Between Generative and Predictive Approach

ABSTRACT

The spread of the SARS-Cov-2 virus in the human race has caused 6.56M deaths worldwide as of Oct 2022 and has brought the economy to a standstill. It has also introduced several challenges worldwide. So, the need of finding a drug that would be able to dilute the symptoms of COVID-19 in the patients. The current methods for drug discovery via conventional methods are a tedious and time-consuming process. So here, the deep learning algorithms come to our rescue. Scientists and Doctors are diligently studying and analyzing the genome sequence of the virus and trying to understand the interaction between the coronavirus protease and a covalent inhibitor. Taking advantage of one such research work published by Shanghai Tech University, the research attempts in making research which is based on an approach to inhibit the protease of SARS-Cov-2(Or any virus) by a covalent inhibitor(also called Ligand). The research was done for some similar viruses to SARS-Cov-2, like SARS, MERS, and HIV. Protein target GI73745819 - SARS Protease, Protein target GI75593047 - HIV pol polyprotein, NS3 - Hep3 protease, and 3CL-Pro - Mers Protease. Bioactivities measured in these papers by medicinal chemists and biochemists are tracked by The National Center for Biotechnology Information (NCBI) which can be accessed by everyone. The goal of this research is to make efforts toward proposing a potentially highly active molecule against a target protein of the 2019 Novel Coronavirus. This research features training of the model in such a way that it predicts the binding power of the drug toward COVID-19 protease. Then compared and reported the inhibition score of ligand and protease to find out one of the best inhibitors. © 2022 IEEE.