Molecular dynamics simulation of complex reactivity with the Rapid Approach for Proton Transport and Other Reactions (RAPTOR) soft- ware package

Gregory Voth,Scott Kaiser,Yuxing Peng,Trung Nguyen,Sijia Chen,Da Teng,Zhi Yue
DOI: https://doi.org/10.26434/chemrxiv-2024-x6031
2024-04-02
Abstract:Simulating chemically reactive phenomena such as proton transport on nanosecond to microsecond and beyond time- scales is a challenging task. Ab initio methods are unable to currently access these timescales routinely, and traditional molecular dynamics methods feature fixed bonding arrangements that cannot account for changes in the system’s bonding topology. The Mul- tiscale Reactive Molecular Dynamics (MS-RMD) method, as implemented in the Rapid Approach for Proton Transport and Other Reactions (RAPTOR) software package for the LAMMPS molecular dynamics code, offers a method to routinely sample longer timescale reactive simulation data with statistical precision. RAPTOR may also be interfaced with enhanced sampling methods to drive simulations towards the analysis of reactive rare events, and a number of collective variables (CVs) have been developed to facilitate this. Key advances to this methodology, including GPU acceleration efforts and novel CVs to model water wire formation are reviewed, along with recent applications of the method which demonstrate its versatility and robustness.
Chemistry
What problem does this paper attempt to address?
This paper attempts to address the challenge of simulating complex reaction phenomena (such as proton transport) on the nanosecond - to - microsecond time scale. Specifically, traditional molecular dynamics methods are unable to capture changes in the bonding topology in the system because they fix the bonding structure, while ab - initio methods, although capable of providing accurate electronic structure calculations, are computationally expensive and difficult to achieve the required long - time - scale simulations. Therefore, the paper proposes a multi - scale reactive molecular dynamics (MS - RMD) method, which combines the advantages of classical molecular dynamics and quantum mechanics and is able to routinely sample reaction simulation data on a longer time scale with statistical accuracy. ### Main problems 1. **Limitations of traditional methods**: - **Ab - initio methods**: Although they can directly calculate the electronic structure, they are computationally expensive and difficult to achieve long - time - scale simulations. - **Standard molecular dynamics**: It uses a force field with a fixed bonding topology and is unable to describe dynamic reaction events. 2. **Requirements for long - time - scale reaction simulations**: - **Proton transport**: For example, proton shuttling through water "wires" or protonatable amino acids in complex systems. - **Protein conformational changes**: How protonation behavior affects protein conformational changes. - **Proton transport in confined spaces**: The proton transport process in a confined environment (such as a proton - exchange membrane). ### Solutions - **Multi - scale reactive molecular dynamics (MS - RMD) method**: - **Multi - state empirical valence bond (MS - EVB) model**: Allows two - state or multi - state on the reaction path to evolve dynamically during the simulation, thus more comprehensively considering the diffusion of protons in the hydrogen - bond network. - **GPU acceleration**: Improves computational efficiency, making long - time - scale simulations possible. - **Enhanced sampling methods**: Such as metadynamics, umbrella sampling, and replica exchange, are used to drive the simulation towards rare - event analysis. - **Collective variables (CVs)**: Develop new CVs to better describe phenomena such as water - wire formation. ### Main contributions of the paper - **Method improvement**: It details the implementation of the MS - RMD algorithm, including state - search algorithms, energy and force calculations, and the diagonalization process. - **Computational optimization**: Proposes multiple computational acceleration techniques, such as domain decomposition, multi - program parallelization, GPU acceleration, and the CGIS method. - **Application examples**: Demonstrates the application of the MS - RMD method in multiple real - world systems, verifying its effectiveness and robustness. Through these improvements, the MS - RMD method is able to simulate complex reaction processes within a reasonable time, providing a powerful tool for studying proton transport and other chemical reactions.