Impact of Timestep on Energy Drift and Accuracy of Alchemical Free Energy Calculations

10 February 2025, Version 1
This content is a preprint and has not undergone peer review at the time of posting.

Abstract

Molecular dynamics (MD) simulations are essential for investigating complex bi- ological systems. Timestep selection is crucial for accuracy and efficiency, yet the common practice of using a 4 fs timestep with hydrogen mass repartitioning (HMR) and SHAKE for alchemical free energy (AFE) calculations requires further scrutiny. This study investigates the impact of timestep on energy drift and AFE calculations across three molecular systems: cAMP, ethane, and the protein-ligand system Tyk2 (emj 42 to emj 55), exploring timesteps from 0.5 to 4 fs. NVE simulations revealed a strong correlation between increasing timestep and energy drift, with and without HMR. Simulations without SHAKE were unstable at larger timesteps. Thermody- namic integration (TI) simulations in the NPT ensemble showed consistent dU dλ values for 0.5, 1, and 2 fs timesteps, but deviations up to 3 kcal/mol were observed at 4 fs in some cases. Statistical t-tests (p < 0.05) confirmed significant differences between the 4 fs timestep and the 0.5 fs reference, especially in aqueous solution. These findings, based on analyses across various λ states (0-1) and softcore treatments, highlight the importance of careful timestep selection. We recommend limiting the timestep to a maximum of 2 fs for accurate and reliable AFE calculations using HMR and SHAKE.

Keywords

Alchemical free energy
times steps

Supplementary materials

Title
Description
Actions
Title
Impact of Timestep on Energy Drift and Accuracy of Alchemical Free Energy Calculations
Description
The total energy drift plots are shown for two systems
Actions

Comments

Comments are not moderated before they are posted, but they can be removed by the site moderators if they are found to be in contravention of our Commenting Policy [opens in a new tab] - please read this policy before you post. Comments should be used for scholarly discussion of the content in question. You can find more information about how to use the commenting feature here [opens in a new tab] .
This site is protected by reCAPTCHA and the Google Privacy Policy [opens in a new tab] and Terms of Service [opens in a new tab] apply.