Reaction Coordinate

Reaction coordinates are low-dimensional representations of complex molecular systems, crucial for simplifying simulations and understanding reaction mechanisms. Current research focuses on developing data-driven methods, employing machine learning architectures like graph neural networks, normalizing flows, and autoencoders, to automatically identify optimal reaction coordinates from molecular dynamics simulations without relying on prior knowledge. These advancements enable more efficient enhanced sampling techniques and improved construction of Markov State Models, ultimately leading to better predictions of molecular kinetics and thermodynamics for diverse systems. This improved understanding has significant implications for fields like drug discovery and materials science.

Papers

September 18, 2024

Graph Neural Network-State Predictive Information Bottleneck (GNN-SPIB) approach for learning molecular thermodynamics and kinetics
Ziyue Zou, Dedi Wang, Pratyush Tiwary
Information Bottleneck Graph Neural Molecular Dynamic GNN Cl Molecular Energy Reaction Coordinate

August 30, 2024

Flow Matching for Optimal Reaction Coordinates of Biomolecular System
Mingyuan Zhang, Zhicheng Zhang, Yong Wang, Hao Wu
Flow Matching Reversible Dynamic Reaction Coordinate

September 11, 2023

Reaction coordinate flows for model reduction of molecular kinetics
Hao Wu, Frank Noé
Model Reduction Molecular System Chemical Kinetics Kinetic Model Reaction Coordinate

February 17, 2023

Enhanced Sampling of Configuration and Path Space in a Generalized Ensemble by Shooting Point Exchange
Sebastian Falkner, Alessandro Coretti, Christoph Dellago
System Configuration Rare Transition Enhanced Sampling Reaction Coordinate

August 22, 2022

Prediction of good reaction coordinates and future evolution of MD trajectories using Regularized Sparse Autoencoders: A novel deep learning approach
Abhijit Gupta
Deep Learning Human Prediction Handwritten Trajectory Molecular Dynamic Sparse Autoencoders Human Reaction Reaction Coordinate

Reaction Coordinate

Papers

Graph Neural Network-State Predictive Information Bottleneck (GNN-SPIB) approach for learning molecular thermodynamics and kinetics

Flow Matching for Optimal Reaction Coordinates of Biomolecular System

Reaction coordinate flows for model reduction of molecular kinetics

Enhanced Sampling of Configuration and Path Space in a Generalized Ensemble by Shooting Point Exchange

Prediction of good reaction coordinates and future evolution of MD trajectories using Regularized Sparse Autoencoders: A novel deep learning approach