Predicting Organic SynthesesKrishna Amin (St Catharine’s). September 10, 2019.
Although we may be somewhat late to the party, we would like to extend our congratulations to Dr Alpha Lee, one of last year’s speakers, and his co-researchers, which included Dr Phillipe Schwaller, at the Cavendish Laboratory with Dr Alpha Lee, and Dr Christopher A. Hunterford and PhD student Peter Bolgar at the Department of Chemistry, for their two remarkable papers reporting on the development of a new model which predicts, with over 90% accuracy, the outcomes of various organic synthesis reactions, the sort that might be involved in medicinal drug discovery. Importantly, this novel model does away with handcrafted rules, which means that the team’s method can be scaled up to cope with the ever-increasing volume of reported organic reactions without significant manhour investment. Furthermore, their method can accurately predict subtle and selective chemical transformations, such as those regarding chemoselectivity and regioselectivity, as well as estimate its own uncertainty.
Lee, A. A.; Yang, Q., Sresht, V.; Bolgar, P., Hou, X.; Klug-McLeod, J. L.; Butler, C. R.. Molecular Transformer Unifies Reaction Prediction and Retrosynthesis across Pharma Chemical Space. Chem. Commun.. 2019, 55, 12152-12155.
Schwaller, P.; Laino, T.; Gaudin, T.; Bolgar, P.; Hunter, C. A.; Bekas, C.; Lee, A. A.. Molecular Transformer: A Model for Uncertainty-Calibrated Chemical Reaction Prediction. ACS Central Science. 2019, 5(9), 1572-1583