Quantum-accurate atomistic modeling of enzyme catalysis using a machine learned potential
Abstract
Electronic rearrangements associated with bond forming/breaking in catalytic enzymes require quantum mechanical (QM) treatment beyond classical molecular mechanics (MM). Hybrid QM/MM methods enable tractable simulations but require system-specific setup and are sensitive to the QM region choice and treatment of the QM/MM interface. We demonstrate quantum-accurate treatment of all-atom, complete enzymes in explicit solvent comprising up to 54k atoms and 1 microsecond of total simulation time using the machine-learned interatomic potential (MLIP) eSEN-omol. We reproduce experimental barrier trends for Claisen rearrangement in chorismate mutase, resolve critical intermediate states in PETase catalyzed polymer depolymerization, and distinguish mechanistic alternatives for metal-activated phosphoryl transfer in nucleoside diphosphate kinase. We realize 1000x speedups relative to typical QM/MM calculations without system-specific tuning. These results establish MLIPs as a practical route to QM-accurate simulations of enzyme catalysis.
Related
- How accurate are foundational machine learning interatomic potentials for heterogeneous catalysis?
- Fast and Accurate Equivariant Foundation Models for Atomistic Simulation
- Assigning the Low-Energy Excitons in Far-Red Photosystem II through Atomistic Modeling
- A Lightweight Universal Machine-Learning Interatomic Potential via Knowledge Distillation for Scalable Atomistic Simulations
- Sequential replica exchange with solute tempering for atomistic modeling of supramolecular polymer structures
BibTeX
@misc{gao2026quantumaccurate,
title = {{Quantum-accurate atomistic modeling of enzyme catalysis using a machine learned potential}},
author = {Meng Gao and Armin Shayesteh Zadeh and Aniruddha Seal and Siva Dasetty and Siddarth K. Achar and Misko Dzamba and Benjamin K. Miller and Leif D. Jacobson and C. Lawrence Zitnick and Brandon M. Wood and Zachary W. Ulissi and Daniel S. Levine and Andrew L. Ferguson},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.09293},
eprint = {2609.09293},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.09293},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Quantum-accurate atomistic modeling of enzyme catalysis using a machine learned potential
AU - Meng Gao
AU - Armin Shayesteh Zadeh
AU - Aniruddha Seal
AU - Siva Dasetty
AU - Siddarth K. Achar
AU - Misko Dzamba
AU - Benjamin K. Miller
AU - Leif D. Jacobson
AU - C. Lawrence Zitnick
AU - Brandon M. Wood
AU - Zachary W. Ulissi
AU - Daniel S. Levine
AU - Andrew L. Ferguson
PY - 2026
DA - 2026/09/16
PB - arXiv
DO - 10.48550/arxiv.2609.09293
UR - https://arxiv.org/abs/2609.09293
AB - Electronic rearrangements associated with bond forming/breaking in catalytic enzymes require quantum mechanical (QM) treatment beyond classical molecular mechanics (MM). Hybrid QM/MM methods enable tractable simulations but require system-specific setup and are sensitive to the QM region choice and treatment of the QM/MM interface. We demonstrate quantum-accurate treatment of all-atom, complete enzymes in explicit solvent comprising up to 54k atoms and 1 microsecond of total simulation time using the machine-learned interatomic potential (MLIP) eSEN-omol. We reproduce experimental barrier trends for Claisen rearrangement in chorismate mutase, resolve critical intermediate states in PETase catalyzed polymer depolymerization, and distinguish mechanistic alternatives for metal-activated phosphoryl transfer in nucleoside diphosphate kinase. We realize 1000x speedups relative to typical QM/MM calculations without system-specific tuning. These results establish MLIPs as a practical route to QM-accurate simulations of enzyme catalysis.
ER -