Improving Few- and Zero-Shot Reaction Template Prediction Using Modern Hopfield Networks

Finding synthesis routes for molecules of interest is essential in the discovery of new drugs and materials. To find such routes, computer-assisted synthesis planning (CASP) methods are employed, which rely on a single-step model of chemical reactivity. In this study, we introduce a template-based single-step retrosynthesis model based on Modern Hopfield Networks, which learn an encoding of both molecules and reaction templates in order to predict the relevance of templates for a given molecule. The template representation allows generalization across different reactions and significantly improves the performance of template relevance prediction, especially for templates with few or zero training examples. With inference speed up to orders of magnitude faster than baseline methods, we improve or match the state-of-the-art performance for top-k exact match accuracy for k ≥ 3 in the retrosynthesis benchmark USPTO-50k. Code to reproduce the results is available at github.com/ml-jku/mhn-react.

PDF Abstract

Datasets


Results from the Paper


Task Dataset Model Metric Name Metric Value Global Rank Benchmark
Single-step retrosynthesis USPTO-50k MHNreact (reaction class unknown) Top-1 accuracy 51.8 # 26
Top-3 accuracy 74.6 # 16
Top-5 accuracy 81.2 # 15
Top-10 accuracy 88.1 # 12
Top-20 accuracy 92.0 # 1
Top-50 accuracy 94.0 # 5

Methods


No methods listed for this paper. Add relevant methods here