Predicting Retrosynthetic Reaction using Self-Corrected Transformer Neural Networks
This work addresses the problem of cumbersome and low-quality computer-aided retrosynthesis for chemists, showing significant improvements in accuracy, especially for novel compounds.
The study tackled retrosynthesis prediction by developing a template-free self-corrected Transformer neural network, achieving 59.0% accuracy on a standard benchmark, which is over 21% higher than other deep learning methods and over 6% higher than template-based methods.
Synthesis planning is the process of recursively decomposing target molecules into available precursors. Computer-aided retrosynthesis can potentially assist chemists in designing synthetic routes, but at present it is cumbersome and provides results of dissatisfactory quality. In this study, we develop a template-free self-corrected retrosynthesis predictor (SCROP) to perform a retrosynthesis prediction task trained by using the Transformer neural network architecture. In the method, the retrosynthesis planning is converted as a machine translation problem between molecular linear notations of reactants and the products. Coupled with a neural network-based syntax corrector, our method achieves an accuracy of 59.0% on a standard benchmark dataset, which increases >21% over other deep learning methods, and >6% over template-based methods. More importantly, our method shows an accuracy 1.7 times higher than other state-of-the-art methods for compounds not appearing in the training set.