Computational Core

The Computational Core serves to

• Aggregate and integrate data from consortium Cores.
• Construct pipelines for automated data processing.
• Disseminate consortium data to the scientific community.

To accomplish this, the Computational Core has developed a Data Repository, and associated software infrastructure. Data is made available to the community via a Web Portal, in two complementary forms: as bulk downloads (both raw and processed data), and via interactive exploration of data - allowing the identification of relationships among the different data types generated by the Cores.

Data Management Overview
A schematic view of the data processing and presentation system (August 2008). The Data Repository serves as the central storage for data. Data generated by the Genomics core are processed through pipelines (developed collaboratively between the Genomics and Computational Cores), and are then loaded into the Data Repository. Likewise, processed data are incorporated from the Reagents, Forward Genetics, Proteomics and Signaling Cores. The Computational Core also provides predictions of possible transcription factor binding sites. Data is made available to the scientific community via a Web Portal. Some features of the of the Web Portal are interactive, in that data can be retrieved dynamically and made accessible through a process of data exploration.

Interactive Data Exploration
The Web Portal allows interactive exploration of data generated by the Cores, along with associated external data. The intention is to provide researchers with the ability to identify relationships between data generated by the various Cores, in order to gain insights that can drive their own research. To begin interactive exploration, one enters the name of any gene in the Integrated Data Browser search box to retrieve the data related to that gene. Interactive exploration will develop over time to include additional data sources and searches.

For any gene or protein of interest, the search provides available

• Gene annotations and links to external and consortium Core data sources.
• Gene expression array profiles.
• Binding sites identified by ChIP-on-chip in the region adjacent to the gene.
• Predicted transcription factor binding sites.

Each of these functionalities is provided in one or more separate web page panels, or "Portlets":

Gene Search
Launches the search for data associated with a given gene.

Gene Summary
Displays, and links out to, associated gene annotations from external and consortium Core data sources.

Gene Expression Studies
Gene expression profiles obtained from consortium 3' expression arrays arrays are displayed graphically and in a table. The user can select any subset of available experimental conditions for display, using a selection box. Expression values are generated using the 3' array processing pipeline.

Genomic Annotations and Features
Provides gene structure information, binding regions identified by ChIP-on-chip studies, and predicted binding sites in the vicinity of the gene. Each feature is listed in terms of mouse genome coordinates. For ChIP-on-chip features, the protein used for the IP is given, as well as pertinent experimental conditions, such as time after cell stimulation. Predicted transcription factor binding sites in the region are obtained from motif scanning. Each potential site is annotated with a specific motif model (Position Weight Matrix) identifier and score indicating the quality of the match.