Learning Molecular Dynamics with Simple Language Model built upon Long Short-Term Memory Neural Network

Abstract

Recurrent neural networks (RNNs) have led to breakthroughs in natural language processing and speech recognition, wherein hundreds of millions of people use such tools on a daily basis through smartphones, email servers and other avenues. In this work, we show such RNNs, specifically Long Short-Term Memory (LSTM) neural networks can also be applied to capturing the temporal evolution of typical trajectories arising in chemical and biological physics. Specifically, we use a character-level language model based on LSTM. This learns a probabilistic model from 1-dimensional stochastic trajectories generated from molecular dynamics simulations of a higher dimensional system. We show that the model can not only capture the Boltzmann statistics of the system but it also reproduce kinetics at a large spectrum of timescales. We demonstrate how the embedding layer, introduced originally for representing the contextual meaning of words or characters, exhibits here a nontrivial connectivity between different metastable states in the underlying physical system. We demonstrate the reliability of our model and interpretations through different benchmark systems and a single molecule force spectroscopy trajectory for multi-state riboswitch. We anticipate that our work represents a stepping stone in the understanding and use of RNNs for modeling and predicting dynamics of complex stochastic molecular systems.

References in corpus (5)

• Canonical sampling through velocity-rescaling
• Double-slit photoelectron interference in strong-field ionization of the neon dimer
• Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting
• Identification of slow molecular order parameters for Markov model construction
• Accurate sampling using Langevin dynamics

Cited by in corpus (9)

• TransPolymer: a Transformer-based language model for polymer property predictions
• Deep learning for dynamic modeling and coded information storage of vector-soliton pulsations in mode-locked fiber lasers
• Forecasting the outcome of spintronic experiments with Neural Ordinary Differential Equations
• Path sampling of recurrent neural networks by incorporating known physics
• Manifold Learning in Atomistic Simulations: A Conceptual Review
• Non-Markovian feedback for optimized quantum error correction
• Super-resolution in Molecular Dynamics Trajectory Reconstruction with Bi-Directional Neural Networks
• Predicting quantum evolutions of excitation energy transfer in a light-harvesting complex using multi-optimized recurrent neural networks
• First-Passage Approach to Optimizing Perturbations for Improved Training of Machine Learning Models