A Neural Network Model Informs Total Synthesis of Clovane Sesquiterpenoids

Abstract

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. Herein, we report 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-endo-trig radical cyclization, and applied to synthetic planning of clovan-2,9-dione, resulting in a 5-step total synthesis that improves on a prior 15-step approach. This work establishes how machine learning can guide multistep syntheses that employ innovative and high-risk human-generated plans.

La revolución quimioinformática: enseñándole química a los robots

“…Esto ha sido demostrado por el grupo de Newhouse en la universidad de Yale al lograr una síntesis concisa de clovan-2,9-diona (Figura 7). 47,48 En el caso de las reacciones de ciclación radicalaria 6-endo-trig, aunque funcionan en determinadas ocasiones, resulta difícil anticipar en qué ocasiones tendrán éxito. Por ello, el grupo de Newhouse entrenó un modelo de inteligencia artificial para predecir el rendimiento de una reacción mediada por una ciclación radicalaria 6-endo-trig.…”

Total Synthesis of Clovan-2,9-dione via [3 + 2 + 1] Cycloaddition and Hydroformylation/Aldol Reaction

“…Unfortunately, this substrate gave both [3 + 2] (major) and [3 + 2 + 1] (minor) products, preventing us from further executing this idea (Scheme e) . The second total synthesis was recently reported by Newhouse, who started from the synthesis of the AC ring and then elegantly used radical cyclization to build the B ring (Scheme f). This route only requires 5 steps to achieve the racemic total synthesis of clovan-2,9-dione.…”