A knowledge-based approach for reaction generation: development, validation and applications
We present a new method for knowledge-based driven reaction generation. The structural changes at the reaction centre are described using reaction vectors1. The reaction vectors are derived automatically from a database of known organic reactions even when no atom-atom mapping information is available. A structure generation algorithm which allows the application of reaction vectors to previously unseen starting materials has been developed. The method reproduces known reactions with a success rate higher than 90%. Its applications in different medicinal chemistry related scenarios and its implementation using Knime2 and ChemAxon’s libraries are discussed.
1 Patel, H., Bodkin, M.J., Chen, B., Gillet, V.J.A Knowledge-Based Approach to De Novo Design Using Reaction Vectors, J. Chem. Inf. Model., 2009, accepted for publication 2 KoNstanz Information MinEr (KNIME), http://www.knime.org/