FlagelLink: a decision support system for distributed flagellar data using data warehouse

Combining different types of data from multiple databases (DBs) is a key feature in bioinformatics, particularly due to the problem that each of these DB resources usually contains different subsets of biological knowledge and only answers questions in its domain, nether helping with questions that span domain boundaries nor considering them. As bioinformatics DBs grow in size and as biological questions grow in scope, better solutions will inevitably consist in preserving the autonomy and diversity of DBs and developing new systems to offer an integrated and transparent access to existing distributed data sources (DS). In this paper, we present a decision support system (DSS), called FlagelLink , to provide access to a set of distributed information about a particular domain (the flagellum, a cellular organelle responsible for motility). It employs useful bioinformatics tools (such as BLAST, MUSCLE, HMMER, etc) in an exclusive data warehouse (DW) through terminology and ontology resources (semantic-driven) to maintain an actual DSS for a specific knowledge domain. FlagelLink (available at http://flagellink.nugen.uece.br/flagellink) has a unified, ondemand integration approach that merges the identified ontological knowledge (which means a defined number of test cases and scenarios of genes and proteins all involved in flagellar activities) with traditional and ontology-based information integration techniques.