Synthetic organic chemistry is one of the cornerstones of the small molecule discovery process and is crucial for rapid navigation of increasingly complex chemical space. The knowledge and experience of researchers has always been the key to combining chemical reactions into successful synthetic schemes. However, chemical synthesis and route design is still a significant challenge. Computer-aided retrosynthesis (CASP) was first proposed by Nobel laureate E J Corey in 1960s. With the rapid advancements of computing hardware and machine learning algorithms, new generation CASP tools hold the promise of bringing an innovative step change in synthesis planning and help chemists plan the strategy for making target molecules. Waller and Segler et al. in collaboration with Reaxys team have developed a deep learning computer algorithm that produces blueprints for the sequences of reactions needed to create small organic molecules, such as drug-like compounds. This has led to an artificial-intelligence tool trained on &gt; 15 million unique single step organic reactions from Reaxys which generates a retrosynthesis tree back to commercially available starting materials for a given target molecule. This approach has the potential to transform the way how synthetic chemists work in the future. We are evaluating the efficacy of this tool, by processing a large variety of different structures, including terpenoids, peptides, heterocyclic and drug-like compounds. To this end, we will present the current capabilities and limitations of the software on selected examples.

Development of Reaxys predictive retrosynthesis, its use and adoption by chemists

ACS Spring 2022

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