Research

Mathematical Sciences

Title :

Understanding the efficacy of existing drug molecules on COVID-19 through an interactive pathway: A deep learning-based predictive model

Area of research :

COVID-19 Research, Life Sciences & Biotechnology, Mathematical Sciences

Focus area :

Mathematical modelling for COVID-19

Principal Investigator :

Prof Rajat Kumar De, Professor, Indian Statistical Institute (ISI), Kolkata

Timeline Start Year :

2020

Contact info :

Details

Executive Summary :

This study aims to adopt/ develop a machine learning (ML) based methodology to predict several new potential drugs for the treatment of SARS-CoV-2.

Outcome/Output:

It was observed that AI-based image processing techniques had a colossal application in the detection of COVID-19 pneumonia in patients, based on chest x-ray, chest computed tomography (CT) and chest high resolution computed tomography (HRTC) images. Further, AI-based predictive models had shown potential in the identification of effective drugs molecules, repurposing of which might help in the treatment of COVID-19 disease. Based on literature reviews and an auto-encoder based deep learning methodology, Mozenavir, Oseltamivir and Di-hydro-artemisinin has been identified as probable drug molecules that might be effective in the treatment of SARS-CoV-2 virus. The available structure of SARS-CoV-2 virus has been analysed and through knowledge-based docking, and identified probable binding sites for vitamin D3 and Ivermectin. It thus opens up new avenues for repurposing of these drug molecules as potential drugs against SARS-CoV-2 viral infection.

Organizations involved