Abstract:Proteins are inherently dynamic, and their conformational ensembles are functionally important in biology. Large-scale motions may govern protein structure–function relationship, and numerous transient but stable conformations of intrinsically disordered proteins (IDPs) can play a crucial role in biological function. Investigating conformational ensembles to understand regulations and disease-related aggregations of IDPs is challenging both experimentally and computationally. In this paper first an unsupervised deep learning-based model, termed Internal Coordinate Net (ICoN), is developed that learns the physical principles of conformational changes from molecular dynamics (MD) simulation data. Second, interpolating data points in the learned latent space are selected that rapidly identify novel synthetic conformations with sophisticated and large-scale sidechains and backbone arrangements. Third, with the highly dynamic amyloid-β (Aβ42) monomer, our deep learning model provided a comprehensive sampling of Aβ42’s conformational landscape. Analysis of these synthetic conformations revealed conformational clusters that can be used to rationalize experimental findings. Additionally, the method can identify novel conformations with important interactions in atomistic details that are not included in the training data. New synthetic conformations showed distinct sidechain rearrangements that are probed by our EPR and amino acid substitution studies. The proposed approach is highly transferable and can be used for any available data for training. The work also demonstrated the ability for deep learning to utilize learned natural atomistic motions in protein conformation sampling.

EquiJump: Protein Dynamics Simulation via SO(3)-Equivariant Stochastic Interpolants

Scalable emulation of protein equilibrium ensembles with generative deep learning

4D Diffusion for Dynamic Protein Structure Prediction with Reference Guided Motion Alignment

Monte Carlo Sampling of Protein Folding by Combining an All-Atom Physics-Based Model with a Native State Bias

Quantum Fragment Based ab Initio Molecular Dynamics for Proteins.

EigenFold: Generative Protein Structure Prediction with Diffusion Models

Accurate calculation of side chain packing and free energy with applications to protein molecular dynamics

Fast Sampling of Protein Conformational Dynamics

JAMUN: Transferable Molecular Conformational Ensemble Generation with Walk-Jump Sampling

Learning Collective Variables with Synthetic Data Augmentation through Physics-Inspired Geodesic Interpolation

Sampling Conformational Ensembles of Highly Dynamic Proteins via Generative Deep Learning

JUMPt: Comprehensive Protein Turnover Modeling of in Vivo Pulse SILAC Data by Ordinary Differential Equations

Fusing Neural and Physical: Augment Protein Conformation Sampling with Tractable Simulations

SE(3)-Stochastic Flow Matching for Protein Backbone Generation

Protein Discovery with Discrete Walk-Jump Sampling

Rapid calculation of side chain packing and free energy with applications to protein molecular dynamics

Utilizing Molecular Dynamics Simulations, Machine Learning, Cryo-EM, and NMR Spectroscopy to Predict and Validate Protein Dynamics

Multiple-Basin Gō-Martini for Investigating Conformational Transitions and Environmental Interactions of Proteins

Understanding Protein Dynamics with L1-Regularized Reversible Hidden Markov Models

Dissipative Particle Dynamics Simulation of Protein Folding in Explicit and Implicit Solvents: Coarse-Grained Model for Atomic Resolution

Navigating protein landscapes with a machine-learned transferable coarse-grained model