Organic synthesis is one of the key stumbling blocks in medicinal chemistry. The necessary yet unsolved step in planning synthesis is solving the forward problem: given reactants and reagents, predict the products. We treat reaction prediction as a machine translation problem between SMILES strings of reactants-reagents and the products. We show that a multi-head attention Molecular Transformer model outperforms all algorithms in the literature, achieving a top-1 accuracy above 90% on a common benchmark dataset. Our algorithm requires no handcrafted rules, and accurately predicts subtle chemical transformations. Crucially, our model can accurately estimate its own uncertainty, with an uncertainty score that is 89\% correct in terms of classifying whether a prediction is correct. Furthermore, we show that model is able to handle inputs without reactant-reagent split and including stereochemistry, which makes our method universally applicable across existing datasets.

Molecular transformer for chemical reaction prediction and uncertainty estimation

2019 Spring National Meeting

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.