Linking biomedical language information and knowledge resources: GO and UMLS.

I. N. Sarkar, M. N. Cantor, R. Gelman, F. Hartel, Y. A. Lussier

Research output: Contribution to journalArticle

19 Scopus citations

Abstract

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.

Original languageEnglish (US)
Pages (from-to)439-450
Number of pages12
JournalPacific Symposium on Biocomputing. Pacific Symposium on Biocomputing
StatePublished - 2003

ASJC Scopus subject areas

  • Medicine(all)

Fingerprint Dive into the research topics of 'Linking biomedical language information and knowledge resources: GO and UMLS.'. Together they form a unique fingerprint.

  • Cite this