Agastya Bhati
Source: https://wsai.iitm.ac.in/faculty/agastya-bhati/ Parent: https://wsai.iitm.ac.in/faculty/
Agastya Bhati
Assistant Professor
agastya.bhati@iitm.ac.inhttps://apbhati1.github.io
Research Interests
AI for Healthcare Human Digital Twins Drug Discovery Mechanistic-AI methods Cancer Multiscale modelling
About
Dr. Agastya P Bhati (Assistant Professor, Wadhwani School of Data Science and AI, IIT Madras) is a computational scientist with extensive experience in the field of computational biomedicine employing cutting-edge scientific methods including those based on artificial intelligence (AI). He completed his BS-MS from the Indian Institute of Science Education and Research (IISER) Mohali in 2014 with Chemical Sciences as his major discipline. Then, he moved to University College London (UCL) for his PhD in the field of computational biomedicine where he developed reliable ensemble-simulation based methods for protein-ligand binding affinity calculations - a key quantity of interest in drug discovery. He employed several enhanced sampling methods in his research and successfully completed his PhD in 2018. He worked as a Scientific Consultant and Researcher in CBK Sci Con Ltd (United Kingdom) for well over a year and then joined as a postdoctoral researcher in the Department of Chemistry at UCL for further research on drug discovery and healthcare. He has an extensive experience of performing large-scale simulations on various (pre)exascale supercomputers across the globe. He is currently leading the development of a novel computational workflow (IMPECCABLE) that combines physics-based methods with those based on AI to accelerate the process of drug discovery by coupling the two in an interactive and iterative way. IMPECCABLE aims at covering all aspects of early-stage drug discovery by not just identifying potent molecules but also optimising other properties such as toxicity and synthesisability. Further, he is active in the development of human digital twins for the advancement of healthcare, making it personalised and predictive. He has published numerous papers in several reputed scientific peer-reviewed journals and is very well cited (You can find him on Google scholar at https://scholar.google.com/citations?user=KodvjxsAAAAJ&hl=en&oi=ao).