Learning to Make Generalizable and Diverse Predictions for Retrosynthesis

Abstract

We propose a new model for making generalizable and diverse retrosynthetic reaction predictions. Given a target compound, the task is to predict the likely chemical reactants to produce the target. This generative task can be framed as a sequence-to-sequence problem by using the SMILES representations of the molecules. Building on top of the popular Transformer architecture, we propose two novel pre-training methods that construct relevant auxiliary tasks (plausible reactions) for our problem. Furthermore, we incorporate a discrete latent variable model into the architecture to encourage the model to produce a diverse set of alternative predictions. On the 50k subset of reaction examples from the United States patent literature (USPTO-50k) benchmark dataset, our model greatly improves performance over the baseline, while also generating predictions that are more diverse.

References in corpus (4)

• Junction Tree Variational Autoencoder for Molecular Graph Generation
• Learning Deep Generative Models of Graphs
• Strategies for Pre-training Graph Neural Networks
• Learning Multimodal Graph-to-Graph Translation for Molecular Optimization

Cited by in corpus (7)

• Learning Graph Models for Retrosynthesis Prediction
• Node-Aligned Graph-to-Graph (NAG2G): Elevating Template-Free Deep Learning Approaches in Single-Step Retrosynthesis
• DirectMultiStep: Direct Route Generation for Multistep Retrosynthesis
• RetCL: A Selection-based Approach for Retrosynthesis via Contrastive Learning
• Data Transfer Approaches to Improve Seq-to-Seq Retrosynthesis
• Permutation invariant graph-to-sequence model for template-free retrosynthesis and reaction prediction
• Modern Hopfield Networks for Few- and Zero-Shot Reaction Template Prediction