A self-learning algorithm for biased molecular dynamics

Abstract

A new self-learning algorithm for accelerated dynamics, reconnaissance metadynamics, is proposed that is able to work with a very large number of collective coordinates. Acceleration of the dynamics is achieved by constructing a bias potential in terms of a patchwork of one-dimensional, locally valid collective coordinates. These collective coordinates are obtained from trajectory analyses so that they adapt to any new features encountered during the simulation. We show how this methodology can be used to enhance sampling in real chemical systems citing examples both from the physics of clusters and from the biological sciences.

6 pages, 5 figures + 9 pages of supplementary information

References in corpus (9)

• Canonical sampling through velocity-rescaling
• Escaping free-energy minima
• Well-Tempered Metadynamics: A Smoothly Converging and Tunable Free-Energy Method
• An efficient, multiple range random walk algorithm to calculate the density of states
• PLUMED: a portable plugin for free-energy calculations with molecular dynamics
• Simulated Tempering: A New Monte Carlo Scheme
• Parallel Tempering Algorithm for Conformational Studies of Biological Molecules
• Sampling rare switching events in biochemical networks
• Action-derived molecular dynamics in the study of rare events

Cited by in corpus (17)

• PLUMED 2: New feathers for an old bird
• Enhanced Sampling in Molecular Dynamics Using Metadynamics, Replica-Exchange, and Temperature-Acceleration
• Enhanced sampling methods for molecular dynamics simulations
• Metadynamics with adaptive Gaussians
• Stress Field Prediction in Cantilevered Structures Using Convolutional Neural Networks
• iMapD: intrinsic Map Dynamics exploration for uncharted effective free energy landscapes
• Recognizing molecular patterns by machine learning: an agnostic structural definition of the hydrogen bond
• Gaussian Mixture Based Enhanced Sampling
• Difference-Based Deep Learning Framework for Stress Predictions in Heterogeneous Media
• Reaction Coordinates for Conformational Transitions using Linear Discriminant Analysis on Positions
• Reaction coordinates and rate constants for liquid droplet nucleation: quantifying the interplay between driving force and memory
• Multistate metadynamics for automatic exploration of conical intersections
• Can molecular dynamics be used to simulate biomolecular recognition?
• Metadynamics for Automatic Sampling of Quantum Property Manifolds: Exploration of Molecular Biradicality Landscapes
• A Sinking Approach to Explore Arbitrary Areas in Free Energy Landscapes
• Using data-reduction techniques to analyse biomolecular trajectories
• Global free energy landscapes as a smoothly joined collection of local maps