Multiscale Neural Network Potential with Anisotropic Message Passing for the Fast and Accurate Simulation of Protein Dynamics and Enzymatic Reactions
- 1Department of Chemistry and Applied Biosciences, ETH Zürich, Vladimir-Prelog-Weg 2, 8093 Zürich, Switzerland
Abstract
We present the next generation of AMP, a neural network potential (NNP) with anisotropic message passing designed to study large biomolecular systems at DFT accuracy in the condensed phase using a multiscale approach similar to quantum-mechanics/molecular-mechanics (QM/MM) with electrostatic embedding. We trained AMPv3 on our recently published biomolecular multiscale simulation (BMS25) data set and demonstrated the model's high efficiency, which enabled us to simulate proteins involving thousands of atoms at DFT accuracy in addition to explicit MM solvent for up to 100 ns, which presents a major leap for contemporary NNPs. We observe excellent scaling to large systems on a single GPU. AMPv3-BMS25 (or AMP-BMS for short) shows promising performance on benchmarks, and we demonstrate that the model can be used to accurately estimate experimental properties, including solvation free energies of small molecules and structural features of proteins. Finally, AMP-BMS/MM was employed to predict the free-energy profiles of reactions catalyzed by the enzymes chorismate mutase and fluoroacetate dehalogenase. In total, AMP-BMS/MM was used to simulate proteins in the condensed phase for a cumulative 23 μs simulation time or 48 billion integration steps. This work establishes AMP-BMS as a highly efficient and accurate model for multiscale simulations of biomolecules.
Related
- Representing chemical and enzymatic reactions in fingerprint space for applicability filtering and classification
- Numerical simulation-guided development of dynamic kinetic resolution reactions
- Population-Specific H-Antigen Chain-Type Variation and Its Potential Impact on Enzymatic Universal-Blood Conversion: A Computational Pilot Study
- Simulation of a Battery Cell on Quantum Computers: Reactions & Transport
- Optimizing Borophene with Copper Foil Support via 3d Transition Metal Loading and Hydroxyl Ligand Modulation: A Potential Catalyst for Oxygen Evolution/Reduction Reactions
BibTeX
@article{thurlemann2026multiscale,
title = {{Multiscale Neural Network Potential with Anisotropic Message Passing for the Fast and Accurate Simulation of Protein Dynamics and Enzymatic Reactions}},
author = {Moritz Thürlemann and Felix Pultar and Igor Gordiy and Enrico Ruijsenaars and Sereina Riniker},
journal = {Journal of the American Chemical Society},
year = {2026},
month = jul,
doi = {10.1021/jacs.6c00217},
url = {https://doi.org/10.1021/jacs.6c00217},
note = {CC BY},
}
RIS
TY - JOUR
TI - Multiscale Neural Network Potential with Anisotropic Message Passing for the Fast and Accurate Simulation of Protein Dynamics and Enzymatic Reactions
AU - Moritz Thürlemann
AU - Felix Pultar
AU - Igor Gordiy
AU - Enrico Ruijsenaars
AU - Sereina Riniker
PY - 2026
DA - 2026/07/01
JO - Journal of the American Chemical Society
DO - 10.1021/jacs.6c00217
UR - https://doi.org/10.1021/jacs.6c00217
AB - We present the next generation of AMP, a neural network potential (NNP) with anisotropic message passing designed to study large biomolecular systems at DFT accuracy in the condensed phase using a multiscale approach similar to quantum-mechanics/molecular-mechanics (QM/MM) with electrostatic embedding. We trained AMPv3 on our recently published biomolecular multiscale simulation (BMS25) data set and demonstrated the model's high efficiency, which enabled us to simulate proteins involving thousands of atoms at DFT accuracy in addition to explicit MM solvent for up to 100 ns, which presents a major leap for contemporary NNPs. We observe excellent scaling to large systems on a single GPU. AMPv3-BMS25 (or AMP-BMS for short) shows promising performance on benchmarks, and we demonstrate that the model can be used to accurately estimate experimental properties, including solvation free energies of small molecules and structural features of proteins. Finally, AMP-BMS/MM was employed to predict the free-energy profiles of reactions catalyzed by the enzymes chorismate mutase and fluoroacetate dehalogenase. In total, AMP-BMS/MM was used to simulate proteins in the condensed phase for a cumulative 23 μs simulation time or 48 billion integration steps. This work establishes AMP-BMS as a highly efficient and accurate model for multiscale simulations of biomolecules.
SN - 0002-7863
SN - 1520-5126
SN - 1943-2984
ER -