http://precedings.nature.com/users/12cd803e743dc2946df152063056ab3f/ Documents posted by William Hogan Nature Publishing Group en Nature Precedings http://precedings.nature.com/images/header_logo.gif http://precedings.nature.com http://dx.doi.org/10.1038/npre.2009.3462.1 IntroductionSeveral researchers have demonstrated that current medical terminologies and ontologies use relations in inconsistent and ambiguous ways,1 despite Woods’ seminal work that first illustrated the problem.2 The goal of the present work is to catalog the different ways in which SNOMED CT uses the is a relation. The rationale for creating the catalog is to serve as a basis for systematically improving the semantics of terminologies and ontologies so as to improve their accessibility to machine inference.MethodsI reviewed the literature to find ontological mistakes that change the interpretation of the is a relation, without respect to any particular terminology. I then reviewed the stated relationships table of SNOMED CT, Jan 2009 version, placing them into categories from the literature, and creating new categories when existing categories did not apply.ResultsI found nine categories of misuse of the is a relation in SNOMED CT (see Table in poster), eight from the literature and one from my analysis. SNOMED CT had an example of every misuse found in the literature.DiscussionThe January 2009 version of SNOMED CT violates its intended interpretation of the is a relation, which is nearly identical to the definition of Smith et al.1 I cataloged nine categories of misuse of is a, and found an example of each in SNOMED CT. This study demonstrates for the first time that (1) certain common ontological mistakes, not previously identified as causing misuse of is a, lead to ambiguity in the interpretation of is a, (2) the stated relationships of SNOMED CT are the source of mistakes in the use of the is a relation, (3) SNOMED CT has at least one example of every problem with is a elucidated from the broader literature.References1. Smith B, Ceusters W, Klagges B, et al. Relations in biomedical ontologies. Genome Biol.2005;6(5):R46.2. Woods W. What’s in a link: Foundations for semantic networks. In: Bobrow D, Collins A, eds. Representation and understanding. New York: Academic Press; 1975:35-82. http://dx.doi.org/10.1038/npre.2009.3462.1 Mon, 27 Jul 2009 14:26:50 UTC What’s in an ‘is a’ link? doi:10.1038/npre.2009.3462.1 2009-07-27 William R. Hogan Nature Precedings 2009-07-27T14:26:50Z Poster Bioinformatics http://creativecommons.org/licenses/by/3.0/ http://dx.doi.org/10.1038/npre.2008.2373.1 Effective translational research requires automated analysis of large datasets collected by multiple researchers working at multiple locations. Reliable, machine interpretation of-—and reasoning with—-large datasets assembled at different times and places by different researchers requires standard representations of data. These representations are controlled, structured vocabularies also known as ontologies. By far, the most successful ontology is the Gene Ontology (GO), used by bioinformatics researchers to annotate genomics data. However, to address the phenotype side of translational research will require annotation of electronic medical record data and clinical research data with a clinical-phenotype ontology analogous to GO. One leading candidate for this ontology is SNOMED-CT (SNCT). However, GO and SNCT are incompatible representations. GO is based on an upper level ontology called Basic Formal Ontology (BFO). In this work, we aligned the upper level of SNCT with BFO to enhance its suitability for translational research. Most (14/19 or 74%) of the top-level concepts of SNCT can be fitted into the framework of BFO, but only after significant reorganization. An important concept that does not align is Clinical Finding, which is intended to comprehend diseases and signs and symptoms of disease. However, a finding of disease (epistemology) is not the same thing as a disease (ontology). This discrepancy between SNCT and BFO is important to consider further. Another key result is that children of the top-level concepts do not necessarily follow their parents into BFO, and thus one must align each SNCT concept independently. Future work is to align the next level of SNCT (345 concepts) with BFO. http://dx.doi.org/10.1038/npre.2008.2373.1 Wed, 08 Oct 2008 09:20:06 UTC Aligning the top-level of SNOMED-CT with Basic Formal Ontology doi:10.1038/npre.2008.2373.1 2008-10-08 William Hogan Nature Precedings 2008-10-08T09:20:06Z Poster Bioinformatics http://creativecommons.org/licenses/by/3.0/