Gene Ontology
Versions (more below) · Term · Attribute · Type
	Ontology Extension Hierarchy
Overview The Gene Ontology, which we abbreviate GO, is an ontology being developed by researchers collaborating in the Gene Ontology Consortium. Currently this involves the three model organism databases: FlyBase (Drosophila), Saccharomyces Genome Database (SGD) and the Mouse Genome Database (MGD). According to Michael Ashburner in the Gene Ontology document, it is hoped that other single-organism databases, such as C. elegans, Arabidopsis and zebra-fish, will join in this collaboration. Eventually, the Gene Ontology will serve the needs of the large community involved in genomic studies. As illustrated by the ontology extension hierarchy on the right, GO extends several other more generic ontologies whose terminologies are used in GO specifications. These more generic ontologies include: the ENZYME Ontology whose structure, as illustrated in the ENZYME - the Enzyme nomenclature database, is based upon the Enzyme Commission number recommendations of the Nomenclature Committee of the International Union of Biochemistry and Molecular Biology (IUBMB), and the LIGAND Compound Ontology whose structure was specified by the LIGAND Compound Database of the Kyoto Encyclopedia of Genes and Genomes (KEGG).
Versions Term This is the IFF version of the GO ontology that is closest in style to the GO Ontology XML file. Term, the only object type, is the genus type of the collection of all GO terms, which are instances of that type. Attribute This IFF version of the GO ontology is in the style of an attribute set for a formal context (local logic). There are four object types: the root type Attribute, and three immediate subtypes Function, Process and Subcellular, which are the genus types of three collections of GO terms. These three collections represent the three aspects or controlled vocabularies of the GO terminology: "molecular function", "biological process" and "subcellular location". Each of the gene expression components "has", "used in" and "occurs in" becomes the incidence (classification) relation of a formal context (local logic), with the collection of all gene products as its object (token) set and the respective controlled vocabulary as its attribute (type) set. Type The IFF version of the GO ontology that is truest to the IFF philosophy and style. Here the "molecular function" controlled vocabulary becomes the type hierarchy for gene products and the "has" component of gene expression becomes the many-to-many type (or instantiation) relation between instances and types. The other two controlled vocabularies, "biological process" and "subcellular location", become theories in the GO ontology and the other two components of gene expression, "used in" and "occurs in", are retained as ontological binary relations in the GO ontology. In one sense, the type version of the GO ontology was constructed from the attribute version of the GO ontology by carefully separating the implicit type structure from the term instances, leaving instances of those new types along with their "part of" relationships encoded as the object-containment structure of those instances.
é
Please send questions, comments and suggestions about this page to: Robert E. Kent rekent@ontologos.org
Copyright © 2000 TOC (The Ontology Consortium). All rights reserved. Revised: July 2000