Login / Signup

Improved Prediction of Molecular Response to Pulling by Combining Force Tempering with Replica Exchange Methods.

Yuvraj SinghGlen M Hocky
Published in: The journal of physical chemistry. B (2024)
Small mechanical forces play important functional roles in many crucial cellular processes, including in the dynamic behavior of the cytoskeleton and in the regulation of osmotic pressure through membrane-bound proteins. Molecular simulations offer the promise of being able to design the behavior of proteins that sense and respond to these forces. However, it is difficult to predict and identify the effect of the relevant piconewton (pN) scale forces due to their small magnitude. Previously, we introduced the Infinite Switch Simulated Tempering in Force (FISST) method, which allows one to estimate the effect of a range of applied forces from a single molecular dynamics simulation, and also demonstrated that FISST additionally accelerates sampling of a molecule's conformational landscape. For some problems, we find that this acceleration is not sufficient to capture all relevant conformational fluctuations, and hence, here we demonstrate that FISST can be combined with either temperature replica exchange or solute tempering approaches to produce a hybrid method that enables more robust prediction of the effect of small forces on molecular systems.
Keyphrases
  • molecular dynamics simulations
  • molecular dynamics
  • single molecule
  • molecular docking
  • single cell
  • machine learning
  • deep learning
  • monte carlo