What

• GenMAPP-CS is being developed as an open-source, Java-based program, incorporating the Cytoscape core to utilize its graph-rendering capabilities, while adding GenMAPP-specific functionality. Our development strategy is two-fold: implementing necessary modifications to the Cytoscape core and adding specialized features for GenMAPP-CS. The Cytoscape program is designed with a highly generalized core, and special features are incorporated as plugins to the core program rather than as part of the core. This design complements the overall implementation strategy of GenMAPP-CS, since many GenMAPP-specific features can be conceptualized as plugins.

  Beyond core developments, GenMAPP-CS will be comprised of a bundle of additional interfaces, tools, and methods that will plug into the fundamental core. Some of these methods may serve as generally useful plugins for Cytoscape, while others will only function in the GenMAPP-CS bundle. Here is where we will implement many of the specific features and functions that make GenMAPP so useful for biologists. Two main interfaces are needed to target the utility of GenMAPP-CS to biologists and the genome-wide analysis of complex datasets: (1) the Database Interface and (2) the Dataset Manager. The Database Interface will link gene objects on a given pathway to known aliases and annotations for those genes, and to the major identifiers used to label microarray datasets. The Dataset Manager will import experimental data in a particular format, applying relevant annotations and filter and managing the analysis and visualization methods. The ability to load, process, and analyze microarray data through intuitive, easy-to-use controls allows biologists to use pathways as research tools and is thus crucial to the success of GenMAPP-CS.

  • 

Why

• Unlike the current VB version, the Java-based GenMAPP-CS program will provide cross-platform support, enhanced functionality, and interoperability. By working with the Cytoscape Consortium, we have identified a unique opportunity to co-develop core visualization and analysis methods for biological networks in a Java, open-source environment. Adopting the architecture of Cytoscape as a starting point for GenMAPP-CS offers clear advantages over rewriting GenMAPP from scratch or attempting to port VB code to Java. Rather than pursuing independent directions in isolation, we can pool our experience and resources to develop a core set of tools that integrate pathway-related information, methods, and data formats. This strategy will provide interoperable tools, establish standards, and promote data exchange, benefiting both developers and the research community.

How

• Specific Aim 1: To build GenMAPP-CS, a client-server version of GenMAPP, to provide a dynamic environment for visualizing and analyzing genomic data on biological pathways. GenMAPP-CS is being developed as an open-source, Java-based program to visualize and analyze datasets that exceed GenMAPP’s current capabilities by 10-100-fold, while maintaining user interfaces and specific functions intuitive to biologists. At the core of GenMAPP-CS will be the architecture and graph-rendering engine of Cytoscape (a generic biological network visualization tool), which we are co-developing to provide GenMAPP-specific functionality. Extensive documentation, application programming interfaces (APIs), and online and hands-on tutorials will be provided for GenMAPP-CS users and open-source developers.

  Specific Aim 2: To dynamically integrate GenMAPP-CS with major gene and pathway databases for over 20 major model organisms. The new GenMAPP-CS architecture will allow us to integrate gene exon, single nucleotide polymorphism (SNP), and protein domain information with probe information at a scale that is impractical in GenMAPP 2.0. The client-server version of GenMAPP will access these data from a local, prepackaged database or from a remote database maintained at GenMAPP.org that will be synchronously updated with major public resources. Active exchange with pubic pathway databases will be facilitated by BioPAX standards, allowing GenMAPP-CS users to access and contribute new pathways.

  Specific Aim 3: To enable GenMAPP-CS to visualize and analyze genome-wide splicing, polymorphism, and interaction datasets. Our laboratory and the GenMAPP user community are actively analyzing whole-genome datasets on transcription, transcript splicing, polymorphisms, and protein-protein interactions. The challenge of analyzing these massive and complex datasets is a major force driving the development of GenMAPP-CS. We have designed several unique analysis methods and visualization strategies for exon-level expression and SNP data. These methods will continue to be developed as modules within GenMAPP-CS.

  • 

  By completing these specific aims, we will dramatically increase the functionality of GenMAPP and its ability to meet the needs of our large, growing user community. In addition, we will merge GenMAPP-CS with the major public pathway resources, providing a direct link that will benefit both the GenMAPP users and the pathway resources. Through our user support efforts for the GenMAPP application, we are in a constant dialogue with the community on how to meet the continuing challenges of pathway analysis. We are confident that we can achieve our goals by building a robust computational environment that will allow us to integrate key resources that synergize with GenMAPP-CS.

•