NIH Postdoctoral Fellow
Synthesis and Catalysis with Predictive Analytics
The synthesis of natural products is motivated by their potent biological effects, enigmatic biosynthetic origins, and complex molecular structures. Computational tools such as density functional theory and machine learning can provide mechanistic insights and predict reactivity in silico to inform synthetic campaigns. In addition, computational tools and predictive analytics can guide the development and application of transition-metal catalysis. Several examples of computational approaches to challenges in synthesis and catalysis are presented: selectivity rationalization during the synthesis of berkeleyone A, machine learning to guide the synthesis of clovane terpenoids, hybrid modeling to predict the site of late-stage functionalization of pharmaceutical intermediates, and machine learning to develop new catalysts for olefin hydroformylation. These advances enable more efficient use of synthetic resources and offer modern solutions to pressing challenges in synthesis and catalysis.