Faculty Bibliography

Integration of various informatics terminologies will be an essential activity towards supporting the advancement of both the biomedical and clinical sciences. The GO consortium has developed an impressive collection of biomedical terms specific to genes and proteins in a variety of organisms. The UMLS is a composite collection of various medical terminologies, pioneered by the National Library of Medicine. In the present study, we examine a variety of techniques for mapping terms from one terminology (GO) to another (UMLS), and describe their respective performances for a small, curated data set attained from the National Cancer Institute, which had precision values ranging from 30% (100% recall) to 95% (74% recall). Based on each technique's performance, we comment on how each can be used to enrich an existing terminology (UMLS) in future studies and how linking biological terminologies to UMLS differs from linking medical terminologies

A major purpose of biomedical terminologies is to provide uniform concept representation, allowing for improved methods of analysis of biomedical information. While this goal is being realized in bioinformatics, with the emergence of the Gene Ontology as a standard, there is still no real standard for the representation of clinical concepts. As discoveries in biology and clinical medicine move from parallel to intersecting paths, standardized representation will become more important. A large portion of significant data, however, is mainly represented as free text, upon which conducting computer-based inferencing is nearly impossible. In order to test our hypothesis that existing biomedical terminologies, specifically the UMLS Metathesaurus and SNOMED CT, could be used as templates to implement semantic and logical relationships over free text data that is important both clinically and biologically, we chose to analyze OMIM (Online Mendelian Inheritance in Man). After finding OMIM entries' conceptual equivalents in each respective terminology, we extracted the semantic relationships that were present and evaluated a subset of them for semantic, logical, and biological legitimacy. Our study reveals the possibility of putting the knowledge present in biomedical terminologies to its intended use, with potentially clinically significant consequences

Human Genome discovery research has set the pace for Post-Genomic Discovery Research. While post-genomic fields focused at the molecular level are intensively pursued, little effort is being deployed in the later stages of molecular medicine discovery research, such as clinical-genomics. The objective of this study is to demonstrate the relevance and significance of integrating mainstream clinical informatics decision support systems to current bioinformatics genomic discovery science. This paper is a feasibility study of an original model enabling novel 'in-silico' clinical-genomic discovery science and that demonstrates its feasibility. This model is designed to mediate queries among clinical and genomic knowledge bases with relevant bioinformatic analytic tools (e.g. gene clustering). Briefly, trait-disease-gene relationships were successfully illustrated using QMR, OMIM, SNOMED-RT, GeneCluster and TreeView. The analyses were visualized as two-dimensional dendrograms of clinical observations clustered around genes. To our knowledge, this is the first study using knowledge bases of clinical decision support systems for genomic discovery. Although this study is a proof of principle, it provides a framework for the development of clinical decision-support-system driven, high-throughput clinical-genomic technologies which could potentially unveil significant high-level functions of genes

Biomedical knowledge is growing at an exponential rate, with new discoveries being published across a range of information sources. A coded, fully-computable, and integrated approach to this information could increase the efficiency of its use, through improved retrieval as well as the eventual ability to apply decision support tools to the knowledge base. Though multiple knowledge bases (KBs) and databases (DBs) concerning gene-disease relationships exist, few present the information in a coded, easily computable form. Focusing on molecular-disease relationships in cancer (gene-disease and protein-disease), we evaluated articles in major biomedical journals, in order to develop both the framework for a knowledge model as well as evaluation criteria. We then used these criteria to evaluate major KBs, DBs, and terminologies. We discovered that although both the high-level as well as the specific molecular-disease relationships present in our test set were mapped in many of the databases, they generally were not applied together in a coded form. We propose a rationale behind a model mediated schema for the integration of these resources