(418b) Similarity Based Functionalization for Enumeration of Synthetically Tractable Chemical Libraries Surrounding a Target

We demonstrate molecular similarity to be an effective metric for proposing single-step functionalization reactions based on analogy to precedent reactions. The developed approach mimics the reaction strategy and suggests co-reactants defined implicitly by a corpus of known reactions. Using ca 348k reactions from the patent literature as a knowledge base, the recorded products or close analogs are among the top 20 proposed products in 74% of ca 44k test reactions. Recursive application of the tool capitalizes on combinatorial chemistry to suggest a large virtual library of analogs for in silico screening. By leveraging common chemistries reported in the patent literature and therefore accessible to most chemists, we have developed a functionalization algorithm capable of enumerating chemical libraries surrounding a target compound of interest.