Predictive Collective Variable Discovery with Deep Bayesian Models

09/18/2018

∙

Extending spatio-temporal scale limitations of models for complex atomistic systems considered in biochemistry and materials science necessitates the development of enhanced sampling methods. The potential acceleration in exploring the configurational space by enhanced sampling methods depends on the choice of collective variables (CVs). In this work, we formulate the discovery of CVs as a Bayesian inference problem and consider the CVs as hidden generators of the full-atomistic trajectory. The ability to generate samples of the fine-scale atomistic configurations using limited training data allows us to compute estimates of observables as well as our probabilistic confidence on them. The methodology is based on emerging methodological advances in machine learning and variational inference. The discovered CVs are related to physicochemical properties which are essential for understanding mechanisms especially in unexplored complex systems. We provide a quantitative assessment of the CVs in terms of their predictive ability for alanine dipeptide (ALA-2) and ALA-15 peptide.

READ FULL TEXT

Predictive Collective Variable Discovery with Deep Bayesian Models

Physics-Constrained Predictive Molecular Latent Space Discovery with Graph Scattering Variational Autoencoder

Embedded-physics machine learning for coarse-graining and collective variable discovery without data

Posterior predictive model checking using formal methods in a spatio-temporal model

Bayesian inference of ODEs with Gaussian processes

Transferable neural networks for enhanced sampling of protein dynamics

URSABench: Comprehensive Benchmarking of Approximate Bayesian Inference Methods for Deep Neural Networks

General Bayesian Inference over the Stiefel Manifold via the Givens Transform

Predictive Collective Variable Discovery with Deep Bayesian Models

Related Research

Physics-Constrained Predictive Molecular Latent Space Discovery with Graph Scattering Variational Autoencoder

Embedded-physics machine learning for coarse-graining and collective variable discovery without data

Posterior predictive model checking using formal methods in a spatio-temporal model

Bayesian inference of ODEs with Gaussian processes

Transferable neural networks for enhanced sampling of protein dynamics

URSABench: Comprehensive Benchmarking of Approximate Bayesian Inference Methods for Deep Neural Networks

General Bayesian Inference over the Stiefel Manifold via the Givens Transform