In phase I of the project we have demonstrated the feasibility of web based morphologic phenotyping of the mouse using magnetic resonance histology. We have exceeded our goal of isotropic resolution at 100 microns by 8x with isotropic resolution at 50 microns. While this spatial resolution provides extraordinary anatomic detail, it results in enormous volumes of data- 1GB/specimen. In this phase II application we address the issues required to deliver these large data sets to the user's desktop via the internet in an efficient, user-friendly fashion. We will lay the groundwork for more extensive use of MR histology for phenotyping the mouse by developing an image database for the ten most critical strains of mice (both male and female specimens) that are currently under study in the National Phenotyping project. The data will be acquired in a standardized fashion and mapped to a standard reference space allowing users to compare any of theses data sets with future data sets from mutant mice in a quantitative, reproducible fashion. We will develop an easily searchable database and interface which will allow the user to select any or all of one or multiple data sets for comparison. We will design plug in modules for Image J, a widely available image display program that will facilitate interaction with these massive image arrays. We will deploy the database, the interface, and the analysis tools making web-based morphologic phenotyping accessible to the broadest possible community. In the second phase of this Phase II effort, we will test the utility of the concept by extending web-based phenotyping to researchers in the Mouse Models of Human Cancer Consortium.
