4.7 Article

Removing Thermostat Distortions of Protein Dynamics in Constant-Temperature Molecular Dynamics Simulations

Journal

JOURNAL OF CHEMICAL THEORY AND COMPUTATION
Volume 17, Issue 9, Pages 5920-5932

Publisher

AMER CHEMICAL SOC
DOI: 10.1021/acs.jctc.1c00448

Keywords

-

Funding

  1. National Institutes of Health [R35 GM118091]

Ask authors/readers for more resources

This study demonstrates that constant-temperature simulations with a Langevin thermostat distort protein dynamics, and presents a correction scheme to remove such distortions. The results show that even for complex systems like proteins with dynamics spanning multiple timescales, one can predict and remove the distortions caused by thermostats.
Molecular dynamics simulations are widely used to determine equilibrium and dynamic properties of proteins. Nearly all simulations, currently, are carried out at constant temperature, with a Langevin thermostat among the most widely used. Thermostats distort protein dynamics, but whether or how such distortions can be corrected has long been an open question. Here, we show that constant-temperature simulations with a Langevin thermostat dilate protein dynamics and present a correction scheme to remove the dynamic distortions. Specifically, ns-scale time constants for overall rotation are dilated significantly but subns time constants for internal motions are dilated modestly, while all motional amplitudes are unaffected. The correction scheme involves contraction of the time constants, with the contraction factor a linear function of the time constant to be corrected. The corrected dynamics of eight proteins are validated by NMR data for rotational diffusion and for backbone amide and side-chain methyl relaxation. The present work demonstrates that even for complex systems like proteins with dynamics spanning multiple timescales, one can predict how thermostats distort protein dynamics and remove such distortions. The correction scheme will have wide applications, facilitating force-field parameterization and propelling simulations to be on par with NMR and other experimental techniques in determining dynamic properties of proteins.

Authors

I am an author on this paper
Click your name to claim this paper and add it to your profile.

Reviews

Primary Rating

4.7
Not enough ratings

Secondary Ratings

Novelty
-
Significance
-
Scientific rigor
-
Rate this paper

Recommended

No Data Available
No Data Available