Automated production of small-molecule dictionaries for use in crystallographic refinements

Many macromolecules are now being studied crystallographically in complexes with a range of ligands and other associated molecules. It is necessary to have templates describing the expected geometry of such molecules before r efinement and model building can be carried out. This paper describes a met hod for generating templates beginning from the SMILES description of the m olecule, the final format of the molecular template being based on the mmCI F definitions for chemical composition. Additionally, the program SMILE2DIC T, which converts the SMILES string to a more extended format, is described . The description details the input required, the output produced and how t he program relates to attempts to automate the procedure of model building for crystallographic refinement. Examples of input to and output from the p rogram are given.