Prof. Seema Mishra

Artificial intelligence/machine learning tools in protein structure prediction: We use AlphaFold and RoseTTAFold in our work on studies on protein-protein interactions and protein-RNA complexes in health and disease. We also use machine learning methods such as ExtraTreeRegressor and Random Forest to study drug-protein target interactions. Computer-aided drug designing, Molecular dynamics to study drug-target interactions: We further use powerful molecular dynamics simulations to understand the structural basis of drug-protein target interactions in cancer occurring inside cells, and exploring the inside views of the consequential domain movements and essential conformational changes with atomistic accuracy. Immunoinformatics in cancer immunotherapy: Applying immunoinformatics approaches to kickstart epitope-based vaccine development in cancers and SARS-CoV-2 has also been one of our frontrunner programs. The powerful approach of Immunoinformatics can be understood from the following observation. From thousands of potential epitopes generated computationally, and out of only two CTL epitopes of SARS-CoV-2 spike protein that were observed to be promiscuous (binding to several MHC class I supertypes) and immunogenic in big data Immunoinformatics studies conducted in March 2020 from my lab as a single author (Royal Society Open Science, September 2020), one CTL epitope (269-YLQPRTFLL-277) has been found generating the strongest CD8+ T cell responses in multiple independent studies using blood samples from convalescent patients (published in PNAS, 2020; Immunity, Nov and Dec 2020). This further reflects the timeliness of results and accuracy of computational tools well corroborated by independent experiments.