Improving ontologies by automatic reasoning and evaluation of logical definitions

BMC Bioinformatics

Software

Georgios V Gkoutos¹  Gabriele Carletti²  Sebastian Bauer³  Sebastian Köhler⁴  Peter N Robinson⁵  Chris J Mungall⁶  Cynthia L Smith⁷  Paul Schofield⁸ 

[1] Department of Genetics, University of Cambridge, Downing Street, Cambridge CB2 3EH, Cambridge, UK;Dipartimento di Matematica e Informatica, Università di Camerino, Via Madonna delle Carceri 9, 62032, Camerino (MC), Italy;Institute for Medical Genetics and Human Genetics, Charité-Universitätsmedizin Berlin, Augustenburger Platz 1, 13353, Berlin, Germany;Institute for Medical Genetics and Human Genetics, Charité-Universitätsmedizin Berlin, Augustenburger Platz 1, 13353, Berlin, Germany;Berlin-Brandenburg Center for Regenerative Therapies (BCRT), Charité-Universitätsmedizin Berlin, Augustenburger Platz 1, 13353, Berlin, Germany;Institute for Medical Genetics and Human Genetics, Charité-Universitätsmedizin Berlin, Augustenburger Platz 1, 13353, Berlin, Germany;Berlin-Brandenburg Center for Regenerative Therapies (BCRT), Charité-Universitätsmedizin Berlin, Augustenburger Platz 1, 13353, Berlin, Germany;Max Planck Institute for Molecular Genetics, Ihnestrasse 73, 14195, Berlin, Germany;Lawrence Berkeley National Laboratory, Mail Stop 64R0121, 94720, Berkeley, CA, USA;The Jackson Laboratory, 04609, Bar Harbor, ME, USA;The Jackson Laboratory, 04609, Bar Harbor, ME, USA;Department of Physiology, Development, and Neuroscience, University of Cambridge, CB2 3EG, Cambridge, UK;

关键词: Gene Ontology;    Biomedical Ontology;    Human Phenotype Ontology;    Gulo;    Logical Definition;   

DOI  :  10.1186/1471-2105-12-418

 received in 2011-06-06, accepted in 2011-10-27,  发布年份 2011

来源: Springer

【 摘 要 】

BackgroundOntologies are widely used to represent knowledge in biomedicine. Systematic approaches for detecting errors and disagreements are needed for large ontologies with hundreds or thousands of terms and semantic relationships. A recent approach of defining terms using logical definitions is now increasingly being adopted as a method for quality control as well as for facilitating interoperability and data integration.ResultsWe show how automated reasoning over logical definitions of ontology terms can be used to improve ontology structure. We provide the Java software package GULO (Getting an Understanding of LOgical definitions), which allows fast and easy evaluation for any kind of logically decomposed ontology by generating a composite OWL ontology from appropriate subsets of the referenced ontologies and comparing the inferred relationships with the relationships asserted in the target ontology. As a case study we show how to use GULO to evaluate the logical definitions that have been developed for the Mammalian Phenotype Ontology (MPO).ConclusionsLogical definitions of terms from biomedical ontologies represent an important resource for error and disagreement detection. GULO gives ontology curators a fast and simple tool for validation of their work.

【 授权许可】

CC BY   
© Köhler et al; licensee BioMed Central Ltd. 2011