The Generation Challenge Programme
     Model-Driven Architecture: Scientific Domain
                Model and Ontology

   Genevieve Mae Aquino1,2 , Jeffrey Detras1 , Rosemary Shrestha2 , Martin
Senger1? , Kevin Manansala1 , Elizabeth Arnaud3 , Reinhard Simon4 , Jayashree
Balaji5 , Thomas Hazekamp6 , Guy Davenport2 , Graham McLaren1 , Theo van
                   Hintum7 , and Richard Bruskiewich11??
     1
       Crop Research Informatics Laboratory - International Rice Research Institute
                                   (IRRI), Philippines
 2
    Crop Research Informatics Laboratory - Centro Internacional de Mejoramiento de
                            Maz y Trigo (CIMMYT), Mexico
                     3
                       Bioversity International, Montpellier, France
                     4
                       Centro Internacional de la Papa (CIP), Peru
5
   International Crops Research Institute for the Semi-Arid Tropics (ICRISAT), India
                         6
                           Bioversity International, Rome, Italy
     7
        Centre for Genetic Resources, Wageningen Universiteit en Researchcentrum
                                (WUR), The Netherlands

    The Generation Challenge Programme (GCP; http://www.generationcp.
org) is a globally-distributed crop research consortium directed toward crop im-
provement through the application of comparative biology and genetic resources
characterization to plant breeding. To achieve interoperability and integration
within the GCP across diverse data types published by project-distributed data
sources and consumed by end-user data analysis tools, the GCP adopted the de-
velopment paradigm of a “model-driven architecture”. At the core of the GCP ar-
chitecture is a scientific domain model, heavily parameterized with GCP-indexed
ontology. The GCP scientific domain model is an object model that encapsulates
key crop science concepts and is documented using Unified Modeling Language.
The GCP-indexed ontology reuses established international standards where
available, converts other publicly-available controlled vocabulary into formally-
managed ontology, and develops novel ontology if no public vocabularies yet
exist. The GCP architecture is being translated into various open source object-
oriented software libraries and into data types for various public semantic web
protocols. The GCP model-driven architecture is specified as a common semantic
framework for the implementation of interoperability and integration of diverse
crop data sets and analysis tools. The architecture and associated open source
software are accessible at http://pantheon.generationcp.org

Key words: crop informatics, agriculture, ontology, domain model


?
     Presenting author.
??
     Corresponding author: r.bruskiewich@cgiar.org