Overview

The heart of BN++ is a comprehensive data model called BioCore that allows to model almost all biochemical processes. In addition, the model can be easily extended to account for novel biochemical concepts and mechanisms. Starting from a UML model of BioCore, we have implemented a C++ framework for rapid application development as well as an equivalent Java class library. Both frameworks enable the users to realize new applications efficiently with a well-tested code basis.
Based on the BioCore data model we have developed and implemented an SQL data warehouse system that integrates data sets from external data sources via importers. Already in the current state the data warehouse represents a comprehensive collection of data integrated from the following external and internal sources:

• Sequence databases: SwissProt, RefSeq
• Pathway databases: KEGG, BioCyc, TransPath
• Protein interaction databases: DIP, MINT, IntAct, HPRD
• Transcription factor databases: TransFac
• Protein annotation databases: InterPro, CAP

Based on the Java class library of BN++ we developed BiNA, a graphical user interface (GUI) and network visualizer that enables the users to carry out complex queries of the BN++ warehouse using an intuitive interface without requiring any knowledge about BioCore, the database internals, or SQL. Using sophisticated graph layout algorithms query results can be easily visualized as graphs or networks in a visually appealing manner. This greatly enhances the usability for those who want to apply BN++ to their own field of research.