Using Reasoning to Guide Annotation with Gene Ontology Terms in GOAT

High-quality annotation of biological data is central to bioinformatics. Annotation using terms from ontologies provides reliable computational access to data. The Gene Ontology (GO), a structured controlled vocabulary of nearly 17,000 terms, is becoming the de facto standard for describing the functionality of gene products. Many prominent biomedical databases use GO as a source of terms for functional annotation of their gene-product entries to promote consistent querying and interoperability. However, current annotation editors do not constrain the choice of GO terms users may enter for a given gene product, potentially resulting in an inconsistent or even nonsensical description. Furthermore, the process of annotation is largely an unguided one in which the user must wade through large GO subtrees in search of terms. Relying upon a reasoner loaded with a DAML+OIL version of GO and an instance store of mined GO-term-to-GO-term associations, GOAT aims to aid the user in the annotati...

• Type: Conference
• Year: 2004
• Where: SIGMOD
• Authors: Michael Bada, Daniele Turi, Robin McEntire, Robert Stevens
• Keywords: Database, GO Terms Users, Large Go Subtrees, SIGMOD 2004, Tedious Annotation Process