Currently, genomics data and data repositories in the public domain are expanding at an explosive pace. The wealth of publicly accessible data is beginning to enable cross-cutting integrative translational bioinformatics studies. However, translational discoveries that could be made by mining such public resources are hampered if they lack standard terminologies to describe diagnoses, diseases, and experimental conditions. For discovery to proceed in the eras of e-science, researchers need tools to enable them to find all the data sets relevant to their area of study-spanning the biological scales from molecular studies to clinical medicine-and bridging the research modalities from high-throughput experiments to clinical trials and medical imaging. The key challenge for enabling such search is the annotation of the various resource elements consistently to identify the biomedical concepts to which they relate. These resource elements range from experimental data sets in public repositories, to records of disease associations of gene products in mutation databases, to entries of clinical-trial descriptions. Creating ontology-based annotations from the metadata in biomedical resources and identifying diagnoses, pathological states, and experimental agents contained in those resources allows indexing of the resources, enabling end users to formulate flexible searches for biomedical data. NCBO has developed a prototype system that is integrated with BioPortal known as Ontrez. Ontrez processes the metadata-annotations of gene expression data sets, descriptions of radiology images, clinical-trial reports, as well as abstracts of Pubmed articles to annotate (or tag) them with terms from appropriate ontologies. Ontrez enables researchers to search for biomedical data (such as genomic data sets, medical images, clinical trials and published papers) that are associated with specific ontology terms.
