Sigmoid Accelerated Molecular Dynamics: An Efficient Enhanced Sampling Method for Biosystems

Abstract

Gaussian accelerated molecular dynamics (GaMD) is recognized as a popular enhanced sampling method for tackling long-standing challenges in biomolecular simulations. Inspired by GaMD, Sigmoid accelerated molecular dynamics (SaMD) is proposed in this work by adding a Sigmoid boost potential to improve the balance between the highest acceleration and accurate reweighting. Compared with GaMD, SaMD extends the accessible time scale and improves the computational efficiency as tested in three tasks. In the alanine dipeptide task, SaMD can produce the free energy landscape with better accuracy and efficiency. In the chignolin folding task, the estimated Gibbs free energy difference can converge to the experimental value ∼30% faster. In the protein–ligand binding task, the bound conformations are closer to the crystal structure with a minimal ligand root-mean-square deviation of 1.7 Å. The binding of the ligand XK263 to the HIV protease is reproduced by SaMD in ∼60% less simulation time.

Flexible Gaussian Accelerated Molecular Dynamics to Enhance Biological Sampling

“…In this work, we showcased its potential use to study glycosylation, but the extra sensibility that the selective GaMD offers can be used to any system of interest, only requiring the list of atoms that one wants to selectively accelerate, e.g., a loop or the flexible termini of a protein. In addition, this methodology could be further expanded by combining it with other reported improvements of the GaMD methodology such as replica exchange GaMD or the use of sigmoid acceleration …”

Exploring Chemical Space Using Ab Initio Hyperreactor Dynamics

“…Since then, several formulations for Δ V ( x ), which do not exhibit the above-mentioned discontinuities, have been proposed and applied to enhance the conformational sampling of biomolecules and enable the investigation of slow processes on computationally feasible time scales. Herein, we will focus on the aMD method and its successors, GaMD by Miao et al, and the recent SaMD approach presented by Zhao et al…”

“…We have investigated alternative reactivity enhancement techniques in the context of the computational nanoreactor approach first introduced by Wang et al 6 in combination with the smooth-step regulated external potential developed in our group 8 to enable reaction space exploration at lower thermostat temperatures. Therefore, we have combined hyperdynamics approaches, namely aMD, 49 GaMD, 50 and SaMD, 51 to elevate the PES with the smooth-step driven virtual piston to keep the molecules confined to the nanoreactor sphere within a new reaction space exploration approach termed ab initio hyperreactor dynamics (HRD).…”

“…These rely either on elevating the PES along a predefined reaction coordinate to decrease reaction barriers locally or modifying the canonical probability distribution of the whole system, eliminating the need for prior knowledge about the involved collective variables (CVs). The latter strategy is used in temperature-based methods like replica exchange/parallel tempering (RE/PT) MD or integrated tempering sampling (ITS), , as well as in bias potential-based methods, e.g., hyperdynamics, , accelerated MD (aMD) and its successors, Gaussian accelerated MD (GaMD), and Sigmoid accelerated MD (SaMD). The hyperdynamics scheme by Voter et al , is based on identifying transition regions and elevating them by a boost potential dependent on the Hessian matrix and approximations thereof.…”

Analyzing Molecular Dynamics Trajectories Thermodynamically through Artificial Intelligence

“…Molecular dynamics (MD) simulations have become an essential tool for comprehending the underlying principles of biomolecular systems at the atomic level. − With rapid development of modern computers, MD simulations can explore biological processes on longer time scales, allowing for a more comprehensive understanding of biological processes such as conformational transitions of macromolecules and protein folding. − By incorporating enhanced-sampling techniques such as adaptive biasing force, , umbrella sampling, metadynamics, and replica-exchange MD, Gaussian/Sigmoid accelerated MD, , the timescale of molecular simulations has been greatly improved, , enabling many intricate biochemical mechanisms to be investigated in silico at the atomistic scale. ,,− Simulating complex biochemical processes, such as protein folding, however, generates a substantial number of structures in molecular trajectories. Analyzing a large number of structures in trajectories and further understanding the mechanisms corresponding to these processes remain a significant challenge.…”