GMOD Chado databanks are widely used by several teams from the French National Institute for Agricultural Research (INRA). As a contractor of the Bio3P team (INRA, Rennes, France), Korilog has developed an open-source J2EE infrastructure to provide end-users with a web-based reporting framework for their genomics Chado databanks.
The framework relies on robust Java technologies (Hibernate, Spring, Velocity and Lucene) for the server-side layer, and on Ajax-enhanced reports for the client-side part.To ensure the system can adapt to extensible Chado schemas, the core of the system relies on a Model Driven Architecture enabling the automatic creation of the databank transaction layers (Hibernate/Spring) using UML schemas.
The entire work has been achieved using Agile-based methodologies involving a close collaborative work between Korilog and three public research teams from INRA (Bio3P and URGI) and Cirad's DAP team. The overall system relies on a set of generic data retrieval and formatting services that can be extended and/or customized using some configuration files and Java codes. A default set of services has been setup to meet the needs of these research teams. As an exemple, here are some screenshots taken from the INRA's GMOD Reports System used by Aphidbase and LepidoDB genomics resources.
The project is owned by INRA and will be soon released on a dedicated SourceForge repository under the CeCILL open-source license.