Abstract:The growing interest in machine learning (ML) tools within chemistry and material science stems from their novelty and ability to predict properties almost as accurately as underlying electronic structure calculations or experiments. Transition path sampling (TPS) offers a practical way to explore transition routes between metastable minima such as conformers and isomers on the multidimensional potential energy surface. However, TPS has historically suffered from the computational cost vs. accuracy trade-off between affordable force-field simulations and expensive high-fidelity quantum mechanical calculations. ML interatomic potentials combined with TPS offer a new approach for the exploration of transition pathways at near-quantum mechanical accuracy, while keeping the computational cost comparable to classical force fields. In this study, we employ the HIP-NN-TS and ANI-1x neural network-based ML potentials, both trained on the ANI-1x dataset of 5 million HCNO structures. We first verify the correctness of our approach by applying it to alanine dipeptide and compare the resulting energy surface and transition paths to the literature. Our findings suggest that proposed approach holds promise for conformational searches, as evidenced by the chemical accuracy (errors ≲ 1 kcal/mol) for thermal molecular dynamics trajectories of alanine dipeptide. While we were able to successfully reconstruct alanine dipeptide’s potential landscape using both HIP-NN-TS and ANI-1x frameworks, we observed that ML models with lower accuracy may still recover additional important conformations. We also find that active learning, augmenting the training data by structures taken from TPS trajectories, improved the accuracy by ~30% with small amounts of additional data. Finally, we evaluated a more intricate case, azobenzene, and reiterated that seemingly simple torsions may bear a challenge for ML potentials and limit their applications in TPS. Inability of HIP-NN-TS to correctly describe the energetics of major rotational pathway in azobenzene isomerization highlights deficiencies of the reference method in describing the electronic degrees of freedom. Our study underscores the importance of domain expertise in selecting physically meaningful pathways for benchmarking ML potentials, especially considering the intricacies of electronic structure in chemical dynamics and non-equilibrium processes.

PhysNet Meets CHARMM: A Framework for Routine Machine Learning / Molecular Mechanics Simulations

Challenges for Machine Learning Force Fields in Reproducing Potential Energy Surfaces of Flexible Molecules

Transfer-Learned Potential Energy Surfaces: Towards Microsecond-Scale Molecular Dynamics Simulations in the Gas Phase at CCSD(T) Quality

Extending the atomic decomposition and many-body representation, a chemistry-motivated monomer-centered approach for machine learning potentials

Modern Semiempirical Electronic Structure Methods and Machine Learning Potentials for Drug Discovery: Conformers, Tautomers, and Protonation States

MolE8: Finding DFT Potential Energy Surface Minima Values from Force-Field Optimised Organic Molecules with New Machine Learning Representations

Machine Learning-Based Enhancements of Empirical Energy Functions: Structure, Dynamics and Spectroscopy of Modified Benzenes

Machine Learning in QM/MM Molecular Dynamics Simulations of Condensed-Phase Systems

Machine learning model for non-equilibrium structures and energies of simple molecules

MLRNet: Combining the Physics-Motivated Potential Models with Neural Networks for Intermolecular Potential Energy Surface Construction.

A neural network potential based on pairwise resolved atomic forces and energies

Structure-Based Sampling And Self-Correcting Machine Learning For Accurate Calculations Of Potential Energy Surfaces And Vibrational Levels

Machine Learning Molecular Dynamics for the Simulation of Infrared Spectra

Challenges and Opportunities for Machine Learning Potentials in Transition Path Sampling: Alanine Dipeptide and Azobenzene Studies

Assessment of Embedding Schemes in a Hybrid Machine Learning/Classical Potentials (ML/MM) Approach

Electrostatic Embedding Machine Learning for Ground and Excited State Molecular Dynamics of Solvated Molecules

Systematic improvement of empirical energy functions in the era of machine learning

Neural Network Based ∆-Machine learning approach efficiently brings the DFT potential energy surface to the CCSD(T) quality: a case for the OH + CH3OH reaction

Machine Learning for Polaritonic Chemistry: Accessing chemical kinetics

Machine learning accelerated photodynamics simulations