Retrosynthesis Prediction with Conditional Graph Logic Network

TL;DR

This paper introduces the Conditional Graph Logic Network, a novel graph neural network-based model for retrosynthesis prediction that improves accuracy and provides interpretability by learning when to apply reaction rules.

Contribution

It presents a new conditional graphical model for retrosynthesis that considers chemical feasibility, along with hierarchical sampling for efficiency, outperforming existing methods.

Findings

Achieves 8.1% improvement over state-of-the-art methods.

Provides interpretable predictions for chemical reactions.

Efficient hierarchical sampling reduces computation costs.

Abstract

Retrosynthesis is one of the fundamental problems in organic chemistry. The task is to identify reactants that can be used to synthesize a specified product molecule. Recently, computer-aided retrosynthesis is finding renewed interest from both chemistry and computer science communities. Most existing approaches rely on template-based models that define subgraph matching rules, but whether or not a chemical reaction can proceed is not defined by hard decision rules. In this work, we propose a new approach to this task using the Conditional Graph Logic Network, a conditional graphical model built upon graph neural networks that learns when rules from reaction templates should be applied, implicitly considering whether the resulting reaction would be both chemically feasible and strategic. We also propose an efficient hierarchical sampling to alleviate the computation cost. While achieving a significant improvement of $8.1\%$ over current state-of-the-art methods on the benchmark dataset, our model also offers interpretations for the prediction.