Overview: openMDR

OpenMDR is a suite of software that provides caGrid-compatible semantic metadata management capabilities. The suite includes four different components:
- MDR Core,
- MDR Query,
- MDR Plugin, and
- MDR Domain Model Generator.
MDR Core - an ISO11179-compliant Metadata registry that stores semantic metadata for a domain or institution. MDR Core is extended from UK cancergrid's cgMDR project (http://cancergrid.org
). MDR Core is capable of storing, versioning and maintaining semantic metadata in the form of Common Data Elements (CDEs). Core provides a web frontend for users to create, edit, deploy and re-use CDEs for new UML model development.
MDR Query - An API and Grid Service that supports federated query and retrieval of semantic metadata from repositories across the Grid, including MDR Core and caBIG caDSR.
MDR Plugin - an Enterprise Architect plug-in that allows a modeler to search for relevant semantic metadata using the MDR Query Grid Service. The modeler then annotates the UML model with the chosen CDEs.
MDR Domain Model Generator - a tool used by a caGrid service developer to generate a caGrid domain model from an annotated UML model. This component provides caGrid compatibility by generating a caGrid domain model, an artifact that is required to create a semantically-annotated caGrid Data Service.
In summary, MDR Core is a semantic repository. MDR Query is a grid service used to search multiple semantic repositories (including MDR Core). UML modelers use MDR Plugin to search for semantic concepts from within Enterprise Architect. caGrid Service Developers use the MDR Domain Model Generator to create caGrid data services.
Each of these projects provides functionality which enables federated semantic metadata annotations to be created and used in Grid Service Registration and Discovery. This will now give caGrid service developers the option of using other semantic metadata management tools in addition to that of the NCI caDSR, EVS and others. Additionally, developers will maintain a fast and agile process for annotating and delivering a strongly typed and semantically anchored grid service into production that is independent from policies concerning the use of particular metadata registries.
How does this all fit in?

Steps
- Create Common Data Elements in Local instance of openMDR - metatadata registry using concepts from Local or remote(NCI) LexEVS instance.
- Create an information model - create logical model, data model and Mapping Object Model to Data Model in Enterprise Architect (EA).
- Annotate logical model with CDE's in Enterprise Architect using openMDR EA plugin. The plugin makes use of the mdr Query Service that queries multiple metadata registries (openMDR, caDSR).
- Export annotated XMI
- Run Domain Model Generator to create domain model file(XML file) with grid service metadata.
- Use Introduce Grid Service Authoring Toolkit that consumes the XMI file and domain model XML file to generate a new grid service.
Based on UK Cancergrid's
cgMDR project






