State-of-the-art augmented NLP transformer models for direct and single-step retrosynthesis

By Igor V. Tetko, Pavel Karpov, Ruud Van Deursen, Guillaume Godin. Nature Communications, 2020

372 citations recorded by the source corpus.

We investigated the effect of different training scenarios on predicting the (retro)synthesis of chemical compounds using text-like representation of chemical reactions (SMILES) and Natural Language Processing (NLP) neural network Transformer architecture. We showed that data augmentation, which is a powerful method used in image processing, eliminated the effect of data memorization by neural networks and improved their performance for prediction of new sequences. This effect was observed when augmentation was used simultaneously for input and the target data simultaneously. The top-5 accuracy was 84.8% for the prediction of the largest fragment (thus identifying principal transformation for classical retro-synthesis) for the USPTO-50k test dataset, and was achieved by a combination of SMILES augmentation and a beam search algorithm. The same approach provided significantly better results for the prediction of direct reactions from the single-step USPTO-MIT test set. Our model achieved 90.6% top-1 and 96.1% top-5 accuracy for its challenging mixed set and 97% top-5 accuracy for the USPTO-MIT separated set. It also significantly improved results for USPTO-full set single-step retrosynthesis for both top-1 and top-10 accuracies. The appearance frequency of the most abundantly generated SMILES was well correlated with the prediction outcome and can be used as a measure of the quality of reaction prediction. Development of algorithms to predict reactant and reagents given a target molecule is key to accelerate retrosynthesis approaches. Here the authors demonstrate that applying augmentation techniques to the SMILE representation of target data significantly improves the quality of the reaction predictions.

Subfields: Computational Theory and Mathematics, Environmental Chemistry, Materials Chemistry

View Paper: https://www.semanticscholar.org/paper/5406e153957dd7a165264da6e6e5d81251997404

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Theory and Mathematics

In Environmental Chemistry

In Materials Chemistry

Citations in the corpus

Improving Few- and Zero-Shot Reaction Template Prediction Using Modern Hopfield Networks (2022, methodology): “In addition, we report results of a single run on the USPTO-full dataset preprocessed by Tetko et al. 20 (see Table S4 in the Supporting Information).”
Geometric deep learning on molecular representations (2021, background): “Moreover, GDL provides novel modeling opportunities, such as data augmentation in low data regimes [29, 30].”
Artificial intelligence in drug development (2025, neutral): “The schematic diagram illustrates the seamless integration of AI-driven software with experimental execution and result analysis within the automated synthesis process.”