Biomedical Data Integration

The volume of data in biomedicine is constantly increasing. A key aspect to address biomedical data integration and semantic interoperability is the use of terminologies and ontologies as a common denominator to structure biomedical data and make them interoperable. However, the great number of developed ontologies and their heterogeneity have raised the problem of ontology matching, which consists in automatically finding mappings (correspondences) between entities of different ontologies. At Stanford, the National Center for Biomedical Ontology has developed BioPortal: a repository of biomedical ontologies which also store and compute mappings between hosted ontologies. At LIRMM, we design state-of-the-art ontology matching tools and we have investigated the use of BioPortal mappings as background knowledge to discover new mappings. The results were promising, however during our experiments, we have noticed that the mappings extracted from the portal are not complete, and contain many redundancies. Therefore, we will work together to improve the quality of these mappings. In addition, ontology matching is more challenging when the ontologies are in different languages. LIRMM is especially interested in French ontologies and terminologies. This project will be the opportunity to collaborate with the NCBO team to design a novel approach that allows to discover multilingual mappings between French-English biomedical ontologies.