Applying the functional abnormality ontology pattern to anatomical functions

• Main Authors: Hoehndorf Robert, Ngonga Ngomo Axel-Cyrille, Kelso Janet
• Format: Article
• Language: English
• Published: BMC 2010-03-01
• Series: Journal of Biomedical Semantics
• Online Access: http://www.jbiomedsem.com/content/1/1/4

<p>Abstract</p> <p>Background</p> <p>Several biomedical ontologies cover the domain of biological functions, including molecular and cellular functions. However, there is currently no publicly available ontology of anatomical functions.</p> <p>Consequently, no explicit relation between anatomical structures and their functions is expressed in the anatomy ontologies that are available for various species. Such an explicit relation between anatomical structures and their functions would be useful both for defining the classes of the anatomy and the phenotype ontologies accurately.</p> <p>Results</p> <p>We provide an ontological analysis of functions and functional abnormalities. From this analysis, we derive an approach to the automatic extraction of anatomical functions from existing ontologies which uses a combination of natural language processing, graph-based analysis of the ontologies and formal inferences. Additionally, we introduce a new relation to link material objects to processes that realize the function of these objects. This relation is introduced to avoid a needless duplication of processes already covered by the Gene Ontology in a new ontology of anatomical functions.</p> <p>Conclusions</p> <p>Ontological considerations on the nature of functional abnormalities and their representation in current phenotype ontologies show that we can extract a skeleton for an ontology of anatomical functions by using a combination of process, phenotype and anatomy ontologies automatically. We identify several limitations of the current ontologies that still need to be addressed to ensure a consistent and complete representation of anatomical functions and their abnormalities.</p> <p>Availability</p> <p>The source code and results of our analysis are available at <url>http://bioonto.de</url>.</p>