Streamlining biological data analysis using BioFlow

Abstract

For several obvious and practical reasons, resources needed for biological data analysis are often geographically distributed and are accessible through the Internet. Such resources usually include data repositories, analysis tools, digital documents, and so on. Such an arrangement warrants sophisticated data and process integration tools in order to design ad hoc higher level applications using these online resources. In this paper we present such a system, called the BioFlow, that exploits recent advances in workflow technology and Internet computing in order to provide support for ad hoc application development by hiding aspects related to the heterogeneity and distributive nature of the resources required by user applications. We introduce the salient features of the BioFlow system, discuss briefly its architecture and implementation issues using simple but real life applications. We demonstrate that the declarative language on which BioFlow is based makes our system quite intuitive, easy to use, effective and efficient for ad hoc application design. The approach taken in BioFlow is somewhat similar to the idea of web services in semantic web computing for biological applications.