Call To Action

The purpose of the proposed web application is to provide rapid and accurate predictions of physicochemical properties, such as solvation energy, solubility, and pKa, for drug-like organic molecules. Traditional physics-based methods like molecular dynamics simulations, though accurate, are slow due to their high computational demands. This project aims to develop a faster, data-driven approach using Graph Neural Networks (GNN) and GPU-based deep learning algorithms. The resulting predictive models will be validated against experimental data and made accessible through a web server, facilitating the pharmaceutical industry's ability to predict properties of new or unknown molecules efficiently. This research has been funded by Science and Engineering Research Board (SERB), Govt. of India (project number: MTR/2021/000859).