Semi-Automatic Composition of Web Services for the Bioinformatics Domain

Web service technology is becoming an increasingly popular way to provide data and computational resources in the biological domain, in part because of the advanced features it provides such as language independence, platform independence, ability to discover services and ease of Web-based programmatic access. As the number of Web services provided by the bioinformatics community continues to grow, the need to use these services in combination is becoming more apparent. Typically, Web services are composed to form a process or workflow using a graphical design tool, such as a BPEL designer. However, for the scientific community, use of these BPEL tools requires a high-level of expertise. In the biological domain, there are some research efforts geared towards providing easier ways to compose biological Web services, most notably the interrelated BioMoby/Taverna projects. So far, the tools developed by such projects are still too difficult to be widely used by the average biologist. Furthermore, they tend to not fully follow Web services standards. Therefore, providing tools that lower the learning curve for Web service composition, while still complying with Web service standards could greatly benefit biologists and bioinformaticians.. In this paper, we describe WS-BioZard, a comprehensive framework for addressing this need by using wizards and semantics to provide a suite of tools that support practical semi-automatic composition of Web services in the biological domain.