Kaiwan Bilal
Research Mentor: Paige Bowling
Mentor Department: Chemistry, LSA
Author(s): Kaiwan Bilal, Paige Bowling
Session: Session 4 (1:00 PM – 1:50 PM)
Presentation Type: Poster 7
Abstract
The rise of high-performance computing in chemical research has led to ongoing studies of simulating drug discovery, using computational molecular dynamics simulations to model protein binding energies. As researchers aim to circumvent disease mutations and improve drug efficacy, multisite lambda dynamics simulations (MSLD) have been used to alchemically model free energy changes. With MSLD, molecular mechanics (MM) methods quickly sample numerous snapshots of the complex over time. This ensemble analysis is particularly useful in accounting for protein-system movement, which can non-trivially impact binding interactions. In contrast, quantum mechanics (QM) approaches to modeling binding energies most accurately account for electrostatic interactions and charge polarizations, but are single-point calculations. For ensemble analysis, QM methods quickly become highly time-consuming and computationally expensive in comparison to MM simulations. In our work, we develop a hybrid QM/MM interface, using QM methods to model the changing substituents and the active site(s) and MM methods for the rest of the unchanging core structure. This integrates QM and MM simulations, drawing from CHARMM and Q-Chem outputs to conduct such electrostatically corrected alchemical free energy simulations. To test our interface, we simulated ligand binding and inhibition of matrix metalloproteinases, a zinc-dependent enzyme family. Matrix metalloproteinases’ complex structure and role in tissue remodeling require representative MM resampling of the complex in varied conformations to illustrate reactivity and binding inhibition. Concurrently, the zinc in matrix metalloproteases creates prevalent metal coordination interactions that require accurate electrostatic modeling of the binding region. Through this work, we hope to establish a publicly distributable QM/MM interface, establishing a seamless QM/MM pathway for future research in drug development and biomolecular function.
