Retrosynthesis Prediction with Local Template Retrieval

TL;DR

This paper introduces RetroKNN, a local template retrieval method that enhances retrosynthesis prediction accuracy by combining neural networks with non-parametric template retrieval, validated on two major benchmarks.

Contribution

The paper presents RetroKNN, a novel non-parametric retrieval approach that improves template-based retrosynthesis prediction by integrating local template retrieval with neural network outputs.

Findings

7.1% top-1 accuracy improvement on USPTO-50K

12.0% top-1 accuracy improvement on USPTO-MIT

Effective combination of neural predictions with local template retrieval

Abstract

Retrosynthesis, which predicts the reactants of a given target molecule, is an essential task for drug discovery. In recent years, the machine learing based retrosynthesis methods have achieved promising results. In this work, we introduce RetroKNN, a local reaction template retrieval method to further boost the performance of template-based systems with non-parametric retrieval. We first build an atom-template store and a bond-template store that contain the local templates in the training data, then retrieve from these templates with a k-nearest-neighbor (KNN) search during inference. The retrieved templates are combined with neural network predictions as the final output. Furthermore, we propose a lightweight adapter to adjust the weights when combing neural network and KNN predictions conditioned on the hidden representation and the retrieved templates. We conduct comprehensive experiments on two widely used benchmarks, the USPTO-50K and USPTO-MIT. Especially for the top-1 accuracy, we improved 7.1% on the USPTO-50K dataset and 12.0% on the USPTO-MIT dataset. These results demonstrate the effectiveness of our method.