Efficient syntheses of complex small molecules often involve speculative experimental approaches. The central challenge of such plans is that experimental evaluation of high-risk strategies is resource intensive, as it entails iterative attempts at unsuccessful strategies. This presentation details a complementary strategy that combines creative human-generated synthetic plans with robust computational prediction of the feasibility of key steps in the proposed synthesis.  A neural network model was developed to predict the outcome of a generally disfavored transformation, the 6-<i>endo</i>-<i>trig</i> radical cyclization, and applied to synthetic planning of clovan-2,9-dione, resulting in a concise synthesis. This work establishes how machine learning can guide multistep syntheses that employ innovative and high-risk human-generated plans.

Neural network model informs total synthesis of complex molecules

ACS Spring 2022

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