Abstract. This paper reports our experience with OWL for the Foundational Model of Anatomy (FMA). We show that converting the FMA from Protégé into OWL DL was possible, with most features of the original FMA captured. The conversion relies on translation and enrichment rules, implemented with flexible options. Unsurprisingly, reasoning with OWL proved to be a real challenge, due to the sheer size and complexity of the FMA. As the entire FMA in OWL DL raised inference problems hard to solve in terms of time and memory, an incremental approach was adopted. A number of various smaller versions that Racer could handle were successfully tested. Some inconsistencies were identified and some classes reclassified. The analysis of the results obtained so far shows the benefits of representing the FMA in OWL and, more generally, the usefulness of DLs reasoning techniques for large-scale biomedical ontologies shared on the Web. 1

Formalisms such as description logics (DL) are sometimes expected to help terminologies ensure compliance with sound ontological principles. The objective of this paper is to study the degree to which one DL-based biomedical terminology (SNOMED CT) complies with such principles. We defined seven ontological principles (for example: each class must have at least one parent, each class must differ from its parent) and examined the properties of SNOMED CT classes with respect to these principles. Our major results are: 31 % of the classes have a single child; 27 % have multiple parents; 51 % do not exhibit any differentiae between the description of the parent and that of the child. The applications of this study to quality assurance for ontologies are discussed and suggestions are made for dealing with multiple inheritance.

Better understanding pathologies-genes relationships requires semantic integration of heterogeneous information distributed in multiple `medical' and `biological' sources. This paper presents an ongoing project that aims at developing an information integration system providing a unified access to biomedical resources. The basic idea is to use for semantic integration the existing knowledge available in standard domain terminologies e.g. GeneOntology^TM, UMLS and databanks e.g. HUGO, GOA. A first tool, BioMeKe, has been achieved in that perspective.

The entities described in the Gene Ontology, (i.e., molecular functions, cellular components and biological processes), often make reference (in their names) to other entities, either from GO or from other ontologies, such as ontologies of chemical entities, cell types and organisms. We developed a method for mapping terms from the Open Biomedical Ontology (OBO) family to GO. We show that 55 % of the 17,250 GO terms include in their names the name of some chemical entity (ChEBI). Our findings are consistent with that of other studies. Additionally, our study provides a quantification of the relations between GO terms and terms from other ontologies. 1

The Gene Ontology (GO) is organized in three allegedly independent hierarchies: molecular functions, biological processes, and cellular components. In this paper, we present an approach based on the Chemical Entities of Biological Interest (ChEBI) ontology to identifying dependence relations in GO, especially relations across hierarchies. Our method is based on the identification of the names of ChEBI entities in GO terms. We distinguish between firstorder dependence relations between GO terms that share a common chemical name, and second-order dependence relations between GO entities whose names include two chemicals that are hierarchically related. Of the 10,516 entities in ChEBI, 26 % were identified in the names of 9,431 GO terms (55 % of all GO terms). A total of 771,302 pairs of related GO terms (first-order associations) were computed. Of these, 44 % correspond to dependence relations across hierarchies. These results were compared to the 8,714 pairs of GO terms identified as dependent by lexical and statistical methods in a previous study (once restricted to GO terms whose names include a ChEBI entity). Of these, 3,932 (45%) were identified as first-order relations, and 937 (11%) as secondorder relations. We show that the two kinds of approaches are complementary. The ChEBI-based is independent of the annotations, allowing even rare dependencies to be identified. Moreover, it takes advantage of the subsumption relations between chemicals in ChEBI, and therefore helps identify second-order dependence relations. This approach can be generalized to other ontologies of chemicals as well as other kinds of ontologies.

Abstract. The functional interpretation of microarray experiments requires computerized methods that exploit similarities between gene products with respect to their Gene Ontology (GO) annotations. While GO already represents taxonomic and meronomic relations, our objective is to identify associative relations across Gene Ontology hierarchies. The expected benefit of this effort is that these additional relations can be used to produce more consistent annotations. As a first step toward this goal, we analyze dependence relations between GO terms, first in a set of 23 gene products involved in enterocyte differentiation and later in GOA. Our approach takes into account ontological, lexical, and statistical aspects of dependence. Preliminary results suggest the interest of combining these three approaches for accurately identifying ontological and functional dependence relations in GO. 1

ven intervals of time. They never exist in full in any single instant. Rather, they unfold themselves through time in the way in which the process of blood circulating through the body unfolds itself in time (Bittner & Smith, 2003). To say that an entity is dependent is to assert that it requires a support from other entities in order to be sustained in existence. There is no Cardiac output without a heart and no Cellular motion without some cell which moves. Independent entities, in contrast, require no support of this sort from other entities. They are the substrates for entities in other categories. Dependence relations can obtain also between dependent entities themselves and we can represent the networks of such relations in graph-theoretical terms (Smith, 1992). At some stage, though, dependence relations must bottom out, in entities which are not themselves dependent on anything else. Two important subspecies of dependence relations linking dependent entities can be distingui

In this article we describe an approach to representing and building ontologies advocated by the Bioinformatics and Medical Informatics groups at the University of Manchester. The hand-crafting of ontologies offers an easy and rapid avenue to delivering ontologies. Experience has shown that such approaches are unsustainable. Description logic approaches have been shown to offer computational support for building sound, complete and logically consistent ontologies. A new knowledge representation language, DAML+ OIL, offers a new standard that is able to support many styles of ontology, from hand-crafted to full logic-based descriptions with reasoning support. We describe this language, the OilEd editing tool, reasoning support and a strategy for the language's use. We finish with a current example, in the Gene Ontology Next Generation (GONG) project, that uses DAML+ OIL as the basis for moving the Gene Ontology from its current hand-crafted, form to one that uses logical descriptions of a concept's properties to deliver a more complete version of the ontology. Copyright # 2003 John Wiley & Sons, Ltd.