Gaussian Accelerated Molecular Dynamics in OpenMM

Abstract

Gaussian accelerated molecular dynamics (GaMD) is a computational technique that provides both unconstrained enhanced sampling and free energy calculations of biomolecules. Here, we present the implementation of GaMD in the OpenMM simulation package and validate it on model systems of alanine dipeptide and RNA folding. For alanine dipeptide, 30 ns GaMD production simulations reproduced free energy profiles of 1000 ns conventional molecular dynamics (cMD) simulations. In addition, GaMD simulations captured the folding pathways of three hyperstable RNA tetraloops (UUCG, GCAA, and CUUG) and binding of the rbt203 ligand to the HIV-1 Tar RNA, both of which involved critical electrostatic interactions such as hydrogen bonding and base stacking. Together with previous implementations, GaMD in OpenMM will allow for wider applications in simulations of proteins, RNA, and other biomolecules.

Deep Boosted Molecular Dynamics (DBMD): Accelerating molecular simulations with Gaussian boost potentials generated using probabilistic Bayesian deep neural network

“…In implicit solvent, AIBMD also identified five low-energy conformational states of alanine dipeptide, including β-sheet centered around (-160°, 150°), PII around (-62°, 140°) and (-90°, 61°), α L around (56°, 34°), and α R around (-70°, -27°) ( Figure 2e ). The 1D and 2D free energy profiles of (Φ, Ψ) calculated from AIBMD simulations were in excellent agreements with previous GaMD simulations performed by AMBER 27 , NAMD 29 , and OpenMM 30 . Therefore, simulations of alanine dipeptide have demonstrated the enhanced sampling capability as well as accuracy of AIBMD for both explicit and implicit solvent systems.…”

“…The 1D and 2D free energy profiles of (Φ, Ψ) calculated from AIBMD simulations were in excellent agreements with previous GaMD simulations performed by AMBER 27 , NAMD 29 , and OpenMM 30 . Therefore, simulations of alanine dipeptide have demonstrated the enhanced sampling capability as well as accuracy of AIBMD for both explicit and implicit solvent systems.…”

“…GaMD has been successfully demonstrated on enhanced sampling of ligand binding, protein folding, protein conformational change, as well as protein-membrane, protein-protein, and protein-nucleic acid interactions 3 . GaMD has been implemented in widely used simulation packages including AMBER 27 , NAMD 29 , OpenMM 30 , GENESIS 31 , and TINKER-HP 32 .…”

Predicting Biomolecular Binding Kinetics: A Review

“…GaMD , is developed to apply a harmonic boost potential to enhance sampling with reduced energetic noise. The boost potential normally exhibits a near Gaussian distribution, which enables proper reweighting of the free energy profiles through cumulant expansion to the second order. , GaMD has been successfully applied to simulate important biomolecular processes, including protein/RNA folding, ,, ligand/protein/RNA binding, ,− and protein conformational changes. ,, However, it remained challenging to accurately predict ligand binding kinetic rates through normal GaMD. , Recently, a “selective GaMD” algorithm, called Ligand GaMD (LiGaMD), , has been developed to allow for more efficient sampling of ligand binding and dissociation processes, which thus allows to accurately predict the ligand binding kinetic rates. For the protein ligand binding system, the system contains ligand L , protein P , and the biological environment E. The system potential energy could be decomposed into the following terms V ( r ) = V P , b ( r P ) + V L , b ( r L ) + V E , b ( r E ) + V P P , n b ( r P ) + V L L , n b …”

“…The boost potential normally exhibits a near Gaussian distribution, which enables proper reweighting of the free energy profiles through cumulant expansion to the second order. 28,29 GaMD has been successfully applied to simulate important biomolecular processes, including protein/RNA folding, 29,108,109 ligand/ protein/RNA binding, 108,110−116 and protein conformational changes. 115,117,118 However, it remained challenging to accurately predict ligand binding kinetic rates through normal GaMD.…”