Integrating Biological Data and Tools with Bis

|The access and exploitation of integrated data repositories and applications is critical for life science. Biologists design protocols that typically rely on complex query pipelines accessing various biological electronic resources (data sources and tools) to consistute data sets for analysis and mining. Integration platforms are needed to allow biologists to acces, manipulate and analyze electronic biological data and thus to support scienti c discovery. The design of integration architectures to support life science addresses speci c issues. Biological resources are highly heterogeneous: not only they di er by their data representation, but they also o er radically di erent query capabilities. In this paper we present the Biological Integration System (BIS) that focuses on the data integration aspects while addressing the integration of query capabilities available at the sources. Typically, biological data sources provide complex query and analysis capabilities such as sequence similarity search engines, that need to Zo e Lacroix is at Arizona State University, P.O. Box 876106, Tempe AZ 85287-6106, USA, Phone: (480) 727 6935, email: [email protected] Omar Boucelma and Mehdi Essid are at the Universit e de Provence, LSIS, 39 rue Joliot-Curie, F-13453 Marseille Cedex 13, France, email: fomar,[email protected] be integrated as well as the data. We introduce the notion of derived wrappers that capture additional capabilities to either compensate capabilities lacking at a source, or to adjust an existing capability in order to make it homogeneous with similar capabilities at other sources. BIS combines a data-driven approach with its mediator with an application-driven approach exploiting Web services. Index Terms|Bioinformatics, Mediation, Heterogeneous Resources, Data Integration, Application Integration, Web.