Deep Learning Collective Variables from Transition Path Ensemble

Abstract

The study of the rare transitions that take place between long lived metastable states is a major challenge in molecular dynamics simulations. Many of the methods suggested to address this problem rely on the identification of the slow modes of the system which are referred to as collective variables. Recently machine learning methods have been used to learn the collective variables as functions of a large number of physical descriptors. Among many such methods Deep Targeted Discriminant Analysis has proven to be useful. This collective variable is built from data harvested in short unbiased simulation in the two basins. Here we enrich the set of data on which the Deep Targeted Discriminant Analysis collective variable is built by adding data coming from the transition path ensemble. These are collected from a number of reactive trajectories obtained using the On-the-fly Probability Enhanced Sampling Flooding method. The collective variables thus trained, lead to a more accurate sampling and faster convergence. The performance of these new collective variables is tested on a number of representative examples.

References in corpus (31)

• Escaping free-energy minima
• PLUMED 2: New feathers for an old bird
• Well-Tempered Metadynamics: A Smoothly Converging and Tunable Free-Energy Method
• Identification of slow molecular order parameters for Markov model construction
• A Novel Path Sampling Method for the Calculation of Rate Constants
• From Metadynamics to Dynamics
• VAMPnets: Deep learning of molecular kinetics
• Time-lagged autoencoders: Deep learning of slow collective variables for molecular kinetics
• 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
• Caliber based spectral gap optimization of order parameters (SGOOP) for sampling complex molecular systems
• Data-Driven Collective Variables for Enhanced Sampling
• Entropy based fingerprint for local crystalline order
• Nonlinear Discovery of Slow Molecular Modes using State-Free Reversible VAMPnets
• A variational conformational dynamics approach to the selection of collective variables in metadynamics
• How Neural Networks Extrapolate: From Feedforward to Graph Neural Networks
• Rare Event Kinetics from Adaptive Bias Enhanced Sampling
• The role of water in host-guest interaction
• From enhanced sampling to reaction profiles
• A microscopic description of acid-base equilibrium
• Gaussian Mixture Based Enhanced Sampling
• Time-Lagged t-Distributed Stochastic Neighbor Embedding (t-SNE) of Molecular Simulation Trajectories
• Finding Efficient Collective Variables: The Case of Crystallization
• Collective Variables for the Study of Crystallization
• Multiscale reweighted stochastic embedding (MRSE): Deep learning of collective variables for enhanced sampling
• Capabilities and Limitations of Time-lagged Autoencoders for Slow Mode Discovery in Dynamical Systems
• Blind search for complex chemical pathways using Harmonic Linear Discriminant Analysis
• Metadynamics of paths
• Enhanced sampling of transition states
• Augmented Harmonic Linear Discriminant Analysis

Cited by in corpus (13)

• A unified framework for machine learning collective variables for enhanced sampling simulations: $\text{mlcolvar}$
• Computing the Committor with the Committor: an Anatomy of the Transition State Ensemble
• Enhanced Sampling in the Age of Machine Learning: Algorithms and Applications
• Descriptors-free Collective Variables From Geometric Graph Neural Networks
• Everything everywhere all at once: a probability-based enhanced sampling approach to rare events
• Learning Collective Variables with Synthetic Data Augmentation through Physics-Inspired Geodesic Interpolation
• Metatensor and metatomic: foundational libraries for interoperable atomistic machine learning
• Spectral Map for Slow Collective Variables, Markovian Dynamics, and Transition State Ensembles
• Unveiling interatomic distances influencing the reaction coordinates in alanine dipeptide isomerization: An explainable deep learning approach
• Investigating the hyperparameter space of deep neural network models for reaction coordinates
• Revealing the Atomistic Mechanism of Rare Events in Molecular Dynamics
• Low-Dimensional Projection of Reactive Islands in Chemical Reaction Dynamics Using a Supervised Dimensionality Reduction Method
• More sophisticated is not always better: comparison of similarity measures for unsupervised learning of pathways in biomolecular simulations