Identifying Transition States of Chemical Kinetic Systems using Network Embedding Techniques

TL;DR

This paper introduces a novel network embedding method tailored for directed graphs to identify transition states in chemical kinetic systems, enhancing understanding of complex reactions through low-dimensional representations.

Contribution

It adapts random walk-based network embedding techniques to directed graphs and applies them to identify transition states in chemical reactions, especially entropic systems.

Findings

Effective dimension reduction for chemical transition state identification

Improved detection of transition states in entropic systems

Method outperforms existing approaches in accuracy

Abstract

Using random walk sampling methods for feature learning on networks, we develop a method for generating low-dimensional node embeddings for directed graphs and identifying transition states of stochastic chemical reacting systems. We modified objective functions adopted in existing random walk based network embedding methods to handle directed graphs and neighbors of different degrees. Through optimization via gradient ascent, we embed the weighted graph vertices into a low-dimensional vector space Rd while preserving the neighborhood of each node. We then demonstrate the effectiveness of the method on dimension reduction through several examples regarding identification of transition states of chemical reactions, especially for entropic systems.