Abstract:
Molecular recognition refers to the specific non-covalent interactions at the molecular level. These interactions include hydrogen bonding, van der Waals forces, hydrophobic interactions, electrostatic forces, etc. They arise from differences in size and shape compatibility, chemical properties, charge differences, and other factors. These interactions play a critical role in comprehending numerous biological processes. Hence, understanding biomolecular recognition has applications across various scientific domains, such as Drug discovery, Enzyme Engineering, Biomedical Research, etc.
The thesis aims to grasp the molecular-level details of relevant biological processes through the utilization of computational methods. We have employed molecular dynamics simulation techniques, aided by enhanced sampling methods, to gain insights into the energetics and mechanistic details of relevant biomolecular events, with a primary focus on protein and RNA-based systems. We have employed metadynamics (well-tempered version), a state-of-art enhanced sampling method to understand the molecular events associated with binding-unbinding of some small molecules (designed by an in-house program) by targeting SARS-CoV-2 virus-based systems (spike protein and RNA-dependent RNA polymerase). These free energy calculations were used to validate some of the potential strong binders to the target mentioned above. A similar approach (multiple walker metadynamics) was further extended to understand the complex free energy landscape associated with protein-protein interactions based on β-catenin: T-cell factor (TCF) based system by utilizing parallel bias version of the metadynamics technique. Understanding this process is essential in finding new therapeutics for cancer and finding new hotspots for drug discovery. The results suggest a folding-assisted unbinding mechanism of TCF, assisted by changes in secondary structural elements triggered by hydrogen bond stabilization. Likewise, the multiple walker version of metadynamics is taken forward to understand the phenomena called “RNA breathing” in a protein-RNA recognition process, where we reveal similar energy local base pair transitions in the bound state of the complex. Further, we have investigated the stereospecific influence of a given amino acid (arginine) on an RNA tetraloop system to provide molecular insights into experimentally validated RNA-based theory on the origin of homochirality. In this specific case, we have utilized a controlled sampling approach by a combination of two enhanced sampling methods (umbrella sampling and parallel bias metadynamics) for efficient sampling.
Moreover, we also modeled and investigated the energetics of the selective ion permeation events through synthetic ion channels in biological membranes to support the experimental observations from collaborative projects. Through the studies, we find that the choice of reaction coordinate/s and specific system-dependent utilization of enhanced sampling methods are essential to reveal the energetics and mechanistic aspects in the biological system of interest.
