Altruistic Metadynamics: Multisystem Biased Simulation

Abstract

We present a new simple extension of multiple walker metadynamics which makes it possible to simulate simultaneously multiple different molecular systems and to predict their free energy surfaces, named Altruistic metadynamics. Similarly to basic metadynamics, it uses a bias potential in the form of hills summed over the simulation. Each system adds a big hill to its "own" bias potential and smaller hills to bias potentials of other systems, hence, each system enhances sampling of other systems. This makes it possible to achieve either faster reaching of the stationary point or higher accuracy of the calculated free energy surfaces. This should be efficient in modeling of series of chemically similar systems, for example, in computational drug screening by metadynamics. The method was tested on model energy surfaces, alanine dipeptide modeled in different force fields and monosaccharides of D-hexopyranose series.

Towards High‐Throughput Modelling of Copper Reactivity Induced by Structural Disorder in Amyloid Peptides

“…To build dimers, two identical monomers were placed at a distance between geometry centres of 2nma nd randomly rotated. [28] Each walker is also termed ar eplica hereafter.T he goal of the multiple walkers is to sample all values of SB in ag iven range (0 % SB % 20) while preventing different walkers from repeating sampling of values that have large degeneracy. The latter bias was built with the recent altruistic metadynamics method.…”

“…Not all of the configurations satisfy ag round-state electron wave function, since convergence is not achieved in as mall amount of configurations produced with the more approximate PBE exchange [a] Te mperature T is 300 K, except in stage 0, in which it was increased from 0t o3 00 Kd uring the first 0.3 ns. [28] to build the bias used in the actual stage on the basis of the histogramso ft he numbero fs alt bridges( the collective variable SB) obtained with all walkersinthe previous stage.The last nanosecond of stage8,t hat is, in absenceo fa ny bias,w as used for RMSD analysis. [28] to build the bias used in the actual stage on the basis of the histogramso ft he numbero fs alt bridges( the collective variable SB) obtained with all walkersinthe previous stage.The last nanosecond of stage8,t hat is, in absenceo fa ny bias,w as used for RMSD analysis.…”

“…The latter bias was built with the recent altruistic metadynamics method. [28] Each walker is also termed ar eplica hereafter.T he goal of the multiple walkers is to sample all values of SB in ag iven range (0 % SB % 20) while preventing different walkers from repeating sampling of values that have large degeneracy.…”

“…In the NPT ensemble, the pressure P was keptc onstant at 1bar.W hen a is reported, the value was used in Equation (3) of ref. [28] to build the bias used in the actual stage on the basis of the histogramso ft he numbero fs alt bridges( the collective variable SB) obtained with all walkersinthe previous stage.The last nanosecond of stage8,t hat is, in absenceo fa ny bias,w as used for RMSD analysis.…”

Toward Fast and Accurate Binding Affinity Prediction with pmemdGTI: An Efficient Implementation of GPU-Accelerated Thermodynamic Integration

“…7-10 Currently, a wide range of alchemical free energy methods provides the most robust and accurate estimates of binding affinities from molecular dynamics simulations. 3,7,11-15 These methods include Thermodynamic Integration (TI) 11,16-18 and Free Energy Perturbation (FEP), 11,19-22 as well as analysis through Bennett’s acceptance ratio and its variants (BAR/MBAR) 23-28 and are augmented with enhanced sampling such as replica exchange molecular dynamics, 29 metadynamics, 30 driven adiabatic free energy dynamics, 31 orthogonal space random walk, 32 adaptive integration, 33 and other methods. 34,35 Advanced alchemical free energy methods for binding affinity prediction 9,10,36 have evolved to the point that they approach quantitative predictive accuracy for ligand-binding affinities, 27,37,38 although much work remains, such as addressing sampling and convergence issues, 27,37 to improve precision so as to afford meaningful comparison with experimental uncertainties.…”