Publication Date

5-2024

Advisor(s) - Committee Chair

Matthew Nee, Kevin Williams, Jeremy Maddox

Degree Program

Department of Chemistry

Degree Type

Master of Science

Abstract

Rapid docking of flexible biological macromolecules remains a significant open challenge in protein structure determination. While rigid docking is relatively simple with toolkits such as TagDock, a key obstacle to rapid flexible docking is the complexity and roughness of the free energy surface associated with protein conformational motion (often termed the many-minima problem), meaning conventional molecular dynamics methods do not effectively sample protein conformations near the interaction complex in accessible timescales. Methods such as metadynamics and replica exchange molecular dynamics exist to ameliorate this obstacle, yet these methods use nonphysical biases or random swaps to enhance sampling. In contrast, high temperature molecular dynamics simulations using simulated annealing offer rapid sampling of a continuous trajectory, biased only by an imposed external temperature. Herein, work is performed to extend the rigid docking toolkit TagDock by implementing a simulated annealing workflow to sample protein conformational motion, extract relevant simulation frames, and perform TagDock analysis, yielding decoy structures as much as 39% closer to the target complex.

Disciplines

Bioinformatics | Chemistry | Life Sciences | Other Chemistry | Physical Chemistry

Share

COinS