Abstract: A structural diagram, in the form of a two-dimensional (2-D) sketch,
remains the most effective portrait of a "small molecule" or chemical reaction.
However, such structural diagrams, as for any other core data, cannot be used
in speech (and should not be used in free text). "Good annotation practice" for
biological databases is to use either consistent and widely recognised
terminology or unique identifiers from a dedicated database to refer to the
molecule of interest. Ideally, scientists should use terminology that is both
pronounceable and meaningful. Thus, a viable solution for a bioinformatician is
to use a definitive controlled vocabulary of biochemical compounds and
reactions, which contains both systematic and common names. In addition,
chemical ontologies provide a means for placing entities of interest into wider
chemical, biological or medical contexts. We present some challenges and
achievements in the standardisation of chemical language in biological
databases, with emphasis on three aspects of annotation: 1. good drawing practice: how to draw unambiguous 2-D diagrams; 2. good naming practice: how to give most appropriate names; and
3. good ontology practice: how to link the entity of interest by
defined logical relationships to other entities.
