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.
We present SMILES-embeddings derived from the internal encoder state of a Transformer [1] model trained to canonize SMILES as a Seq2Seq problem. Using a CharNN [2] architecture upon the embeddings results in higher quality interpretable QSAR/QSPR models on diverse benchmark datasets including regression and classification tasks. The proposed Transformer-CNN method uses SMILES augmentation for training and inference, and thus the prognosis is based on an internal consensus. That both the augmentation and transfer learning are based on embeddings allows the method to provide good results for small datasets. We discuss the reasons for such effectiveness and draft future directions for the development of the method. The source code and the embeddings needed to train a QSAR model are available on https ://githu b.com/bigch em/trans forme r-cnn. The repository also has a standalone program for QSAR prognosis which calculates individual atoms contributions, thus interpreting the model's result. OCHEM [3] environment (https ://ochem .eu) hosts the on-line implementation of the method proposed.
<div><div><div><p>We describe a Transformer model for a retrosynthetic reaction prediction task. The model is trained on 45 033 experimental reaction examples extracted from USA patents. It can successfully predict the reactants set for 42.7% of cases on the external test set. During the training procedure, we applied different learning rate schedules and snapshot learning. These techniques can prevent overfitting and thus can be a reason to get rid of internal validation dataset that is advantageous for deep models with millions of parameters. We thoroughly investigated different approaches to train Transformer models and found that snapshot learning with averaging weights on learning rates minima works best. While decoding the model output probabilities there is a strong influence of the temperature that improves at T=1.3 the accuracy of models up to 1-2%.</p></div></div></div>
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.