EcoCyc, like RiboWeb, uses an ontology to describe the richness and complexity of a domain and the constraints acting within that domain, to specify a database schema [15]. EcoCyc is presented to biologists using an encyclopaedia metaphor. It covers E. coli. genes, metabolism, regulation and signal transduction, which a biologist can explore and use to visualise information [16]. The knowledge base currently describes 4391 E. coli. genes, 695 enzymes encoded by a subset of these genes, 904 metabolic reactions and the organisation of these reactions into 129 metabolic pathways. EcoCyc uses the classification of gene product function from Riley [17] as part of this description. Scientists can visualise the layout of genes within the E. coli. chromosome, or of an individual biochemical reaction, or of a complete biochemical pathway (with compound structures displayed).
EcoCyc's use of an ontology to define a database schema has the advantages of its expressivity and ability to evolve quickly to account for the rapid schema changes needed for biological information [15]. The user is not aware of this use of an ontology, except that the constraints expressed in the knowledge captured mean that the complexity of the data held is captured precisely. In EcoCyc, for example, the concept of Gene is represented by a class with various attributes, that link through to other concepts: Polypeptide product, Gene name, synonyms and identifiers used in other databases etc. The representation system can be used to impose constraints on those concepts and instances which may appear in the places described within the system.