Browse by author
Lookup NU author(s): Dr Ioan-Bogdan MagdăuORCiD, Dr Daniel ColeORCiD
This work is licensed under a Creative Commons Attribution 4.0 International License (CC BY 4.0).
Classical empirical force fields have dominated biomolecular simulations for over 50 years. Although widely used in drug discovery, crystal structure prediction, and biomolecular dynamics, they generally lack the accuracy and transferability required for first-principles predictive modeling. In this paper, we introduce MACE-OFF, a series of short-range transferable force fields for organic molecules created using state-of-the-art machine learning technology and first-principles reference data computed with a high level of quantum mechanical theory. MACE-OFF demonstrates the remarkable capabilities of short-range models by accurately predicting a wide variety of gas- and condensed-phase properties of molecular systems. It produces accurate, easy-to-converge dihedral torsion scans of unseen molecules as well as reliable descriptions of molecular crystals and liquids, including quantum nuclear effects. We further demonstrate the capabilities of MACE-OFF by determining free energy surfaces in explicit solvent as well as the folding dynamics of peptides and nanosecond simulations of a fully solvated protein. These developments enable first-principles simulations of molecular systems for the broader chemistry community at high accuracy and relatively low computational cost.
Author(s): Kovács DP, Moore JH, Browning NJ, Batatia I, Horton JT, Pu Y, Kapil V, Witt WC, Magdău IB, Cole DJ, Csányi G
Publication type: Article
Publication status: Published
Journal: Journal of the American Chemical Society
Year: 2025
Volume: 147
Issue: 21
Pages: 17598–17611
Print publication date: 28/05/2025
Online publication date: 19/05/2025
Acceptance date: 02/05/2025
Date deposited: 19/05/2025
ISSN (print): 0002-7863
ISSN (electronic): 1520-5126
Publisher: American Chemical Society
URL: https://doi.org/10.1021/jacs.4c07099
DOI: 10.1021/jacs.4c07099
Data Access Statement: The data used to train the models are publicly available at: 10. 17863/CAM.107498. The torsion drive data set is also available at: https://zenodo.org/records/11385284. The MACE-OFF series of models is available at: https://github. com/ACEsuit/mace-off/tree/main.
Altmetrics provided by Altmetric
