Enhanced Sampling in the Age of Machine Learning: Algorithms and Applications

Abstract

Molecular dynamics simulations hold great promise for providing insight into the microscopic behavior of complex molecular systems. However, their effectiveness is often constrained by long timescales associated with rare events. Enhanced sampling methods have been developed to address these challenges, and recent years have seen a growing integration with machine learning techniques. This review provides a comprehensive overview of how they are reshaping the field, with a particular focus on the data-driven construction of collective variables. Furthermore, these techniques have also improved biasing schemes and unlocked novel strategies via reinforcement learning and generative approaches. In addition to methodological advances, we highlight applications spanning different areas such as biomolecular processes, ligand binding, catalytic reactions, and phase transitions. We conclude by outlining future directions aimed at enabling more automated strategies for rare-event sampling.

References in corpus (95)

• Escaping free-energy minima
• Variational Inference: A Review for Statisticians
• PLUMED 2: New feathers for an old bird
• Well-Tempered Metadynamics: A Smoothly Converging and Tunable Free-Energy Method
• On representing chemical environments
• Machine learning and the physical sciences
• A Data-Driven Approximation of the Koopman Operator: Extending Dynamic Mode Decomposition
• PLUMED: a portable plugin for free-energy calculations with molecular dynamics
• Statistically optimal analysis of samples from multiple equilibrium states
• Identification of slow molecular order parameters for Markov model construction
• A high-bias, low-variance introduction to Machine Learning for physicists
• Physics-inspired structural representations for molecules and materials
• VAMPnets: Deep learning of molecular kinetics
• Enhanced Sampling in Molecular Dynamics Using Metadynamics, Replica-Exchange, and Temperature-Acceleration
• Time-lagged autoencoders: Deep learning of slow collective variables for molecular kinetics
• Forward Flux Sampling for rare event simulations
• Simulating rare events in equilibrium or non-equilibrium stochastic systems
• Rethinking Metadynamics: from bias potentials to probability distributions
• Enhanced sampling methods for molecular dynamics simulations
• A Variational Approach to Enhanced Sampling and Free Energy Calculations
• Elaborating Transition Interface Sampling Methods
• Caliber based spectral gap optimization of order parameters (SGOOP) for sampling complex molecular systems
• Data-driven approximation of the Koopman generator: Model reduction, system identification, and control
• Data-driven model reduction and transfer operator approximation
• Guided Image Generation with Conditional Invertible Neural Networks
• Molecular enhanced sampling with autoencoders: On-the-fly collective variable discovery and accelerated free energy landscape exploration
• Data-Driven Collective Variables for Enhanced Sampling
• Solving Statistical Mechanics Using Variational Autoregressive Networks
• Variational Encoding of Complex Dynamics
• Unsupervised machine learning in atomistic simulations, between predictions and understanding
• Nonlinear Discovery of Slow Molecular Modes using State-Free Reversible VAMPnets
• A Unified Approach to Enhanced Sampling
• Variational Koopman models: slow collective variables and molecular kinetics from short off-equilibrium simulations
• Neural networks-based variationally enhanced sampling
• A variational conformational dynamics approach to the selection of collective variables in metadynamics
• Autonomous artificial intelligence discovers mechanisms of molecular self-organization in virtual experiments
• State Predictive Information Bottleneck
• Enhancing entropy and enthalpy fluctuations to drive crystallization in atomistic simulations
• Reinforced dynamics for enhanced sampling in large atomic and molecular systems
• Targeted free energy perturbation
• Neural network based path collective variables for enhanced sampling of phase transformations
• Identification of simple reaction coordinates from complex dynamics
• A unified framework for machine learning collective variables for enhanced sampling simulations: $\text{mlcolvar}$
• Rare Event Kinetics from Adaptive Bias Enhanced Sampling
• Efficient sampling of high-dimensional free energy landscapes using adaptive reinforced dynamics
• From enhanced sampling to reaction profiles
• The role of water in host-guest interaction
• Targeted free energy estimation via learned mappings
• Navigating at Will on the Water Phase Diagram
• Computing Committor Functions for the Study of Rare Events Using Deep Learning
• Computing the Committor with the Committor: an Anatomy of the Transition State Ensemble
• Learning Free Energy Landscapes Using Artificial Neural Networks
• Folding a Small Protein Using Harmonic Linear Discriminant Analysis
• Deep Learning Collective Variables from Transition Path Ensemble
• From data to noise to data: mixing physics across temperatures with generative artificial intelligence
• Time-Lagged t-Distributed Stochastic Neighbor Embedding (t-SNE) of Molecular Simulation Trajectories
• Free energy surface reconstruction from umbrella samples using Gaussian process regression
• Unfolding Hidden Barriers by Active Enhanced Sampling
• Uncertainty-biased molecular dynamics for learning uniformly accurate interatomic potentials
• Girsanov reweighting for path ensembles and Markov state models
• Finding Efficient Collective Variables: The Case of Crystallization
• Enhanced Sampling in the Age of Machine Learning: Algorithms and Applications
• Explaining reaction coordinates of alanine dipeptide isomerization obtained from deep neural networks using Explainable Artificial Intelligence (XAI)
• Driving and characterizing nucleation of urea and glycine polymorphs in water
• Collective Variables for the Study of Crystallization
• Multiscale reweighted stochastic embedding (MRSE): Deep learning of collective variables for enhanced sampling
• Predictive Collective Variable Discovery with Deep Bayesian Models
• Integrated VAC: A robust strategy for identifying eigenfunctions of dynamical operators
• Skipping the Replica Exchange Ladder with Normalizing Flows
• Reaction Coordinates for Conformational Transitions using Linear Discriminant Analysis on Positions
• How wet should be the reaction coordinate for ligand unbinding?
• Capabilities and Limitations of Time-lagged Autoencoders for Slow Mode Discovery in Dynamical Systems
• Descriptors-free Collective Variables From Geometric Graph Neural Networks
• Reaction coordinates and rate constants for liquid droplet nucleation: quantifying the interplay between driving force and memory
• Characterizing metastable states with the help of machine learning
• Everything everywhere all at once: a probability-based enhanced sampling approach to rare events
• Computing committors via Mahalanobis diffusion maps with enhanced sampling data
• PySAGES: flexible, advanced sampling methods accelerated with GPUs
• Spectral Map: Embedding Slow Kinetics in Collective Variables
• Timewarp: Transferable Acceleration of Molecular Dynamics by Learning Time-Coarsened Dynamics
• Structure and polymerization of liquid sulfur across the $\lambda$-transition
• Multimap targeted free energy estimation
• Optimal reaction coordinates and kinetic rates from the projected dynamics of transition paths
• Is the Local Ion Density Sufficient to Drive NaCl Nucleation from the Melt and Aqueous Solution?
• Understanding the role of predictive time delay and biased propagator in RAVE
• Efficient Approximation of Molecular Kinetics using Random Fourier Features
• Solving eigenvalue PDEs of metastable diffusion processes using artificial neural networks
• Autoencoders for dimensionality reduction in molecular dynamics: collective variable dimension, biasing and transition states
• Advanced simulations with PLUMED: OPES and Machine Learning Collective Variables
• Learning Collective Variables with Synthetic Data Augmentation through Physics-Inspired Geodesic Interpolation
• Efficient mapping of phase diagrams with conditional Boltzmann Generators
• Spectral Map for Slow Collective Variables, Markovian Dynamics, and Transition State Ensembles
• Learning Markovian Dynamics with Spectral Maps
• A Boltzmann generator for the isobaric-isothermal ensemble
• Self-Supervised Evolution Operator Learning for High-Dimensional Dynamical Systems

Cited by in corpus (4)

• Enhanced Sampling in the Age of Machine Learning: Algorithms and Applications
• Deep learning of committor for ion dissociation and interpretable analysis of solvent effects using atom-centered symmetry functions
• Unraveling the Mechanism of Drug Binding to SARS-CoV-2 RNA Pseudoknot with Thermodynamics-Driven Machine Learning
• Deep learning of committor and explainable artificial intelligence analysis for identifying reaction coordinates