Without good analytical tools, interpreting microarray data will remain a slow and arduous process. We will develop microarray analysis software to accelerate discovery of causal or diagnostic relationships between gene expression and genetic, environmental, pathological, and developmental conditions. This software will integrate user-configurable analysis tools, multiple visualization methods, and direct interactions with external databases, to help scientists interpret microarray data. These tools will be provided in a readily-extensible architecture so data from other gene expression detection techniques and newly discovered analytical methods can be easily incorporated. Finally, this software will help automate repetitive analytical steps, thus speeding the transition from data to hypotheses. This software will facilitate any research or diagnosis that employs microarrays, including clinical studies, drug discovery, and basic research. During Phase 1, we will design and implement software prototypes that fulfills these functional requirements: 1. Microarray analysis software must have a robust and extensible mathematical analysis architecture. Our architecture will allow users to create their own analytical methods as well as employ those that come packaged with the software. 2. It must have a robust and extensible visualization architecture. The user must be presented with an intuitive and informative interface for selecting from the available microarray experiments, and he should be able to group experiments for subsequent analyses and to share with collaborators. 3. It must provide user-configurable tools to automate data curation and analysis.
