VAMPnets: Deep learning of molecular kinetics

Abstract

There is an increasing demand for computing the relevant structures, equilibria and long-timescale kinetics of biomolecular processes, such as protein-drug binding, from high-throughput molecular dynamics simulations. Current methods employ transformation of simulated coordinates into structural features, dimension reduction, clustering the dimension-reduced data, and estimation of a Markov state model or related model of the interconversion rates between molecular structures. This handcrafted approach demands a substantial amount of modeling expertise, as poor decisions at any step will lead to large modeling errors. Here we employ the variational approach for Markov processes (VAMP) to develop a deep learning framework for molecular kinetics using neural networks, dubbed VAMPnets. A VAMPnet encodes the entire mapping from molecular coordinates to Markov states, thus combining the whole data processing pipeline in a single end-to-end framework. Our method performs equally or better than state-of-the art Markov modeling methods and provides easily interpretable few-state kinetic models.

References in corpus (2)

• Variational cross-validation of slow dynamical modes in molecular kinetics
• Statistically optimal analysis of state-discretized trajectory data from multiple thermodynamic states

Cited by in corpus (98)

• Machine Learning for Fluid Mechanics
• Machine learning and the physical sciences
• Machine Learning Force Fields
• Deep learning for universal linear embeddings of nonlinear dynamics
• Machine learning for molecular simulation
• Data-driven discovery of coordinates and governing equations
• Towards Exact Molecular Dynamics Simulations with Machine-Learned Force Fields
• Enhancing Computational Fluid Dynamics with Machine Learning
• Time-lagged autoencoders: Deep learning of slow collective variables for molecular kinetics
• Autonomous discovery in the chemical sciences part I: Progress
• Data Driven Governing Equations Approximation Using Deep Neural Networks
• Autonomous discovery in the chemical sciences part II: Outlook
• SINDy-PI: A Robust Algorithm for Parallel Implicit Sparse Identification of Nonlinear Dynamics
• sGDML: Constructing Accurate and Data Efficient Molecular Force Fields Using Machine Learning
• Data-Driven Collective Variables for Enhanced Sampling
• Deep Neural Networks with Koopman Operators for Modeling and Control of Autonomous Vehicles
• Perspective on integrating machine learning into computational chemistry and materials science
• Stress Field Prediction in Cantilevered Structures Using Convolutional Neural Networks
• Variational Encoding of Complex Dynamics
• Coarse-Graining Auto-Encoders for Molecular Dynamics
• Nonlinear Discovery of Slow Molecular Modes using State-Free Reversible VAMPnets
• Sparse Identification of Nonlinear Dynamics for Rapid Model Recovery
• Graph Dynamical Networks for Unsupervised Learning of Atomic Scale Dynamics in Materials
• Molecular Force Fields with Gradient-Domain Machine Learning: Construction and Application to Dynamics of Small Molecules with Coupled Cluster Forces
• Training neural nets to learn reactive potential energy surfaces using interactive quantum chemistry in virtual reality
• Physics-constrained, low-dimensional models for MHD: First-principles and data-driven approaches
• Strategies for the Construction of Machine-Learning Potentials for Accurate and Efficient Atomic-Scale Simulations
• Galerkin Approximation of Dynamical Quantities using Trajectory Data
• SchNetPack 2.0: A neural network toolbox for atomistic machine learning
• A unified framework for machine learning collective variables for enhanced sampling simulations: $\text{mlcolvar}$
• Special Topic: Markov Models of Molecular Kinetics
• Ensemble Learning of Coarse-Grained Molecular Dynamics Force Fields with a Kernel Approach
• A Perspective on Deep Learning for Molecular Modeling and Simulations
• Deep Learning Collective Variables from Transition Path Ensemble
• GraphVAMPNet, using graph neural networks and variational approach to markov processes for dynamical modeling of biomolecules
• Challenges for Machine Learning Force Fields in Reproducing Potential Energy Surfaces of Flexible Molecules
• 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)
• Deep Learning Enhanced Dynamic Mode Decomposition
• Ten Quick Tips for Deep Learning in Biology
• Deep Learning Models for Global Coordinate Transformations that Linearize PDEs
• Ensuring thermodynamic consistency with invertible coarse-graining
• Integrated VAC: A robust strategy for identifying eigenfunctions of dynamical operators
• Capabilities and Limitations of Time-lagged Autoencoders for Slow Mode Discovery in Dynamical Systems
• Coarse-graining Molecular Systems by Spectral Matching
• Learning forecasts of rare stratospheric transitions from short simulations
• MSM/RD: Coupling Markov state models of molecular kinetics with reaction-diffusion simulations
• A Deep Autoencoder Framework for Discovery of Metastable Ensembles in Biomacromolecules
• Neural Canonical Transformation with Symplectic Flows
• A kernel-based approach to molecular conformation analysis
• Kernel methods for detecting coherent structures in dynamical data
• Characterizing metastable states with the help of machine learning
• Structure-preserving Method for Reconstructing Unknown Hamiltonian Systems from Trajectory Data
• Multiscale molecular kinetics by coupling Markov state models and reaction-diffusion dynamics
• Spectral Map: Embedding Slow Kinetics in Collective Variables
• Data-driven Computation of Molecular Reaction Coordinates
• Fast conformational clustering of extensive molecular dynamics simulation data
• Multidimensional minimum-work control of a 2D Ising model
• Manifold Learning in Atomistic Simulations: A Conceptual Review
• Coupling rare event algorithms with data-based learned committor functions using the analogue Markov chain
• Koopman analysis of quantum systems
• Deeptime: a Python library for machine learning dynamical models from time series data
• Efficient Approximation of Molecular Kinetics using Random Fourier Features
• Deep learning the slow modes for rare events sampling
• Artificial Intelligence for Science in Quantum, Atomistic, and Continuum Systems
• Optimization of Non-Equilibrium Self-Assembly Protocols Using Markov State Models
• Algorithmic Differentiation for Automated Modeling of Machine Learned Force Fields
• Deflation reveals dynamical structure in nondominant reaction coordinates
• Non-equilibrium Markov state modeling of periodically driven biomolecules
• Exploring glassy dynamics with Markov state models from graph dynamical neural networks
• Construction of Machine Learned Force Fields with Quantum Chemical Accuracy: Applications and Chemical Insights
• Elucidating Interfacial Dynamics of Ti-Al Systems Using Molecular Dynamics Simulation and Markov State Modeling
• tgEDMD: Approximation of the Kolmogorov Operator in Tensor Train Format
• Accurate Molecular Dynamics Enabled by Efficient Physically-Constrained Machine Learning Approaches
• Error bounds for dynamical spectral estimation
• Learning Markovian Dynamics with Spectral Maps
• Extended dynamic mode decomposition with dictionary learning using neural ordinary differential equations
• Dynamical systems and complex networks: A Koopman operator perspective
• Early warning indicators via latent stochastic dynamical systems
• Hierarchical deep learning-based adaptive time-stepping scheme for multiscale simulations
• Relevant, hidden, and frustrated information in high-dimensional analyses of complex dynamical systems with internal noise
• Unveiling interatomic distances influencing the reaction coordinates in alanine dipeptide isomerization: An explainable deep learning approach
• Learning Interpretable Collective Variables for Spreading Processes on Networks
• Investigating the hyperparameter space of deep neural network models for reaction coordinates
• Collective variables between large-scale states in turbulent convection
• Recovering Hidden Degrees of Freedom Using Gaussian Processes
• Methods to Recover Unknown Processes in Partial Differential Equations Using Data
• From reductionism to realism: Holistic mathematical modelling for complex biological systems
• Learning dynamical systems from data: Gradient-based dictionary optimization
• Efficient spectra from atomistic simulation: a generalized master equation study of the air-water interface
• Revealing the Atomistic Mechanism of Rare Events in Molecular Dynamics
• Kalman Filter Aided Federated Koopman Learning
• Data-Driven Time Propagation of Quantum Systems with Neural Networks
• Improving Molecular Force Fields Across Configurational Space by Combining Supervised and Unsupervised Machine Learning
• Markov State Models for Tracking Reaction Dynamics on Catalytic Nanoparticles
• Deep learning of committor and explainable artificial intelligence analysis for identifying reaction coordinates
• Improving Estimation of the Koopman Operator with Kolmogorov-Smirnov Indicator Functions
• Information thermodynamics of transition paths between multiple mesostates