Integrated bioinformatics application for automated target discovery
Article first published online: 19 JAN 2005
Copyright © 2005 Wiley Periodicals, Inc.
Journal of the American Society for Information Science and Technology
Special Issue: Bioinformatics
Volume 56, Issue 5, pages 483–492, March 2005
How to Cite
Toldo, L. and Rippmann, F. (2005), Integrated bioinformatics application for automated target discovery. J. Am. Soc. Inf. Sci., 56: 483–492. doi: 10.1002/asi.20137
- Issue published online: 9 FEB 2005
- Article first published online: 19 JAN 2005
- Manuscript Accepted: 22 APR 2004
In this article we present an in silico method that automatically assigns putative functions to DNA sequences. The annotations are at an increasingly conceptual level, up to identifying general biomedical fields to which the sequences could contribute. This bioinformatics data-mining system makes substantial use of several resources: a locally stored MEDLINE® database; a manually built classification system; the MeSH® taxonomy; relational technology; and bioinformatics methods. Knowledge is generated from various data sources by using well-defined semantics, and by exploiting direct links between them. A two-dimensional “Concept Map™” displays the knowledge graph, which allows causal connections to be followed. The use of this method has been valuable and has saved considerable time in our in-house projects, and can be generally exploited for any sequence-annotation or knowledge-condensation task.