Reweighted autoencoded variational Bayes for enhanced sampling (RAVE).

By João Marcelo Lamim Ribeiro, P. Bravo, Yihang Wang, Pratyush Tiwary. 2018

320 citations recorded by the source corpus.

Here we propose the reweighted autoencoded variational Bayes for enhanced sampling (RAVE) method, a new iterative scheme that uses the deep learning framework of variational autoencoders to enhance sampling in molecular simulations. RAVE involves iterations between molecular simulations and deep learning in order to produce an increasingly accurate probability distribution along a low-dimensional latent space that captures the key features of the molecular simulation trajectory. Using the Kullback-Leibler divergence between this latent space distribution and the distribution of various trial reaction coordinates sampled from the molecular simulation, RAVE determines an optimum, yet nonetheless physically interpretable, reaction coordinate and optimum probability distribution. Both then directly serve as the biasing protocol for a new biased simulation, which is once again fed into the deep learning module with appropriate weights accounting for the bias, the procedure continuing until estimates of desirable thermodynamic observables are converged. Unlike recent methods using deep learning for enhanced sampling purposes, RAVE stands out in that (a) it naturally produces a physically interpretable reaction coordinate, (b) is independent of existing enhanced sampling protocols to enhance the fluctuations along the latent space identified via deep learning, and (c) it provides the ability to easily filter out spurious solutions learned by the deep learning procedure. The usefulness and reliability of RAVE is demonstrated by applying it to model potentials of increasing complexity, including computation of the binding free energy profile for a hydrophobic ligand-substrate system in explicit water with dissociation time of more than 3 min, in computer time at least twenty times less than that needed for umbrella sampling or metadynamics.

Subfields: Artificial Intelligence, Materials Chemistry, Molecular Biology

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Materials Chemistry

In Molecular Biology

Citations in the corpus

Deep Generative Markov State Models (2018, neutral): “Most of these methods primarily aim at obtaining a low-dimensional latent space that encodes the long-time behavior of the system, and the decoder takes the role of defining or regularizing the learning problem [30, 8, 14, 19, 23].”
Enhanced Sampling Methods for Molecular Dynamics Simulations [Article v1.0] (2022, methodology): “Among methods that we can categorise as ABP methods are local elevation [107], energy landscape paving [108], selfhealing umbrella sampling [109, 110], adaptive biasing MD (ABMD) [111], Gaussian-mixture umbrella sampling [112], the adaptive biasing potential method [113], basis function sampling [11”
Machine learning for molecular simulation (2019, methodology): “Machine learning has also been used in combination with enhanced sampling methods to learn the free energy surface on-the-fly [26, 60, 61, 22, 62, 63, 64, 65].”
Unsupervised Learning Methods for Molecular Simulation Data (2021, background): “…constructing biased dynamics in the learned manifold space and reweighting the resulting simulations to obtain equilibrium thermodynamic quantities, 162,164 − 166 as well as by selecting starting points for unbiased MD simulations which can then be similarly reweighted to obtain equilibrium…”
Generating valid Euclidean distance matrices (2019, background): “Sampling different geometric structures is, however, still largely left to molecular dynamics (MD) simulation that suffers from the rare event sampling problem, although recently machine learning has been used to speed up MD simulation [23, 24, 25, 26, 27] or to perform sampling of the equilibrium d”
From sequence to Boltzmann weighted ensemble of structures with AlphaFold2-RAVE (2022, unknown): “SPIB belongs to the “Reweighted Autoencoded Variational Bayes for Enhanced Sampling (RAVE)” family of methods 10,11 .”