My Research

What do I do?

I am a Computational Chemist/Biologist with specialization in computational modeling using information from various biophysical/biochemical techniques. My core expertise lies in modelling, designing and optimizing biomolecular complexes using Molecular Dynamics and Artificial Intelligence based approaches.

Postdoctoral Research

Studying evolving biomolecular systems using solution scattering

I have recently developed a chemometrics based method called COSMiCS to objectively decompose large solution scattering datasets to retrieve both the SAXS profiles and the concentrations of the components constituting the evolving system. Chemometrics is defined as the chemical discipline that uses mathematical, statistical, and other methods employing formal logic to design or select optimal measurement procedures and experiments, and to provide maximum relevant chemical information by analyzing chemical data. What distinguishes chemometrics from pure statistics is that appropriate constraints are applied so that the chemically and physically sensible solutions are derived. By evolving system, I refer to any biomolecular system that gradually moves from state A to state B. Our method can be used to study a vast variety of polydisperse biomolecular systems.

Integrative Modelling of flexible biomolecules

My other area of interest is to develop ways to integrate data from different experimental techniques to model ensembles that fit all the data. . A notable example is our study of the highly flexible complex between von Willebrand factor (vWF) and ADAMTS13. The two proteins are crucial components of blood homeostasis system where vWF, which exists as large multimers, helps in the formation of platelet plugs at the site of injury. ADAMTS13 is a metalloprotease whose only function is to control the platelet adhesion activity of vWF by cleaving it to smaller forms. The fact that nature has dedicated an enzyme solely for the cleavage of vWF emphasizes the relevance of this system. I wrote a parallel genetic algorithm that distributes the task of evaluating various combinations of conformations to multiple processors. This allowed us to filter through a search space of more than a trillion combinations to select sub-ensembles that simultaneously justify eleven intermolecular crosslinks derived from quantitative cross-linking/Mass spectrometry studies and five protected surfaces derived from H/D exchange studies. The integrative analysis of the structural data reveals how multiple transient interactions between the disordered vWF and flexible ADAMTS13 result in a high-affinity complex and lead to specific protease activity.