Progress in deep Markov State Modeling: Coarse graining and experimental data restraints
arXiv:2108.01927 · doi:10.1063/5.0064668
Abstract
Recent advances in deep learning frameworks have established valuable tools for analyzing the long-timescale behavior of complex systems such as proteins. Especially the inclusion of physical constraints, e.g. time-reversibility, was a crucial step to make the methods applicable to biophysical systems. Furthermore, we advance the method by incorporating experimental observables into the model estimation showing that biases in simulation data can be compensated for. We further develop a new neural network layer in order to build an hierarchical model allowing for different level of details to be studied. Finally, we propose an attention mechanism which highlights important residues for the classification into different states. We demonstrate the new methodology on an ultralong molecular dynamics simulation of the Villin headpiece miniprotein.
References in corpus (20)
- Adam: A Method for Stochastic Optimization
- PyTorch: An Imperative Style, High-Performance Deep Learning Library
- A Unified Approach to Interpreting Model Predictions
- Show, Attend and Tell: Neural Image Caption Generation with Visual Attention
- Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization
- Methods for Interpreting and Understanding Deep Neural Networks
- Identification of slow molecular order parameters for Markov model construction
- Interpretable Explanations of Black Boxes by Meaningful Perturbation
- From Metadynamics to Dynamics
- Time-lagged autoencoders: Deep learning of slow collective variables for molecular kinetics
- Learning how to explain neural networks: PatternNet and PatternAttribution
- Bayesian ensemble refinement by replica simulations and reweighting
- Molecular enhanced sampling with autoencoders: On-the-fly collective variable discovery and accelerated free energy landscape exploration
- Variational cross-validation of slow dynamical modes in molecular kinetics
- Projected and Hidden Markov Models for calculating kinetics and metastable states of complex molecules
- Nonlinear Discovery of Slow Molecular Modes using State-Free Reversible VAMPnets
- Graph Dynamical Networks for Unsupervised Learning of Atomic Scale Dynamics in Materials
- Interpreting CNNs via Decision Trees
- Deep learning Markov and Koopman models with physical constraints
- Deflation reveals dynamical structure in nondominant reaction coordinates