Data-Driven Chemistry for Developing Organic Synthesis Routes for Functional Chemicals

We have been investigating a method of developing synthesis routes for functional chemicals by combining a synthetic route design system (SRDS) and theoretical calculations. This method called the data-driven synthesis route development, consists of (i) creating synthetic routes by SRDS and (ii) verifying their feasibility on the basis of transition state (TS) searches using QC calculations. We adopt the AIPHOS/TOSP developed by Funatsu et al. as the SRDS in our NEDO project. TOSP uses Transforms of the knowledge-based database of name reactions. It has been considered very difficult to find TSs without knowledge of chemical reactions and quantum chemistry. The TS motif method we developed led to solve this problem. The method takes into account the similarity of TSs, which makes quick searches for new TSs feasible. Furthermore, constructing a database of TS motifs makes it possible to handle in a short time many synthesis routes SRDS creates. We applied the digital method to develop synthesis route of a drug, Branebrutinib.