PAG-XI  Plant & Animal Genomes XI Conference

January 11-15, 2003
Town & Country Convention Center
San Diego, CA

Workshop: Bioinformatics Interfaces, Ontologies and Interoperability
            

W65

MIPS GENOME DATABASES - INTERFACES, ONTOLOGIES AND INTEROPERABILITY

Heiko Schoof

MIPS/IBI, GSF National Research Center of Environment and Health, Ingolstaedter Landstr. 1, 85764 Neuherberg, Germany

With the MIPS Arabidopsis thaliana genome database (MAtDB) our aim is to go beyond a data repository towards creating an integrated knowledge resource. To this end, the Arabidopsis genome has been a backbone against which to structure and integrate heterogenous data. The challenges are:

  1. Keeping data current
  2. Designing data models that can evolve with new data
  3. Facilitating integration of external data
  4. Enabling comprehensive views of the data
  5. Providing simple access to the data for analysis tools
  6. Transferring knowledge to other species
Towards this end we require simple yet powerful interfaces and a flexible database system. At MIPS, we opted for a federated approach rather than a data warehouse, and a number of databases and systems form the backend. This will not be restricted to local databases, a Europe-wide network of plant databases is the aim of the PlaNet project. Beside the web interface, some annotation and maintenance tools have been implemented, like a semi-automatic update of gene models based on cDNA data or an XML-based system for rapid integratin of expert annotation on protein families. For a comprehensive, genome-wide view, catalogs like the MIPS functional catalog, gene ontology or phenotype catalog are helpful. The next step is to use the comprehensive data from model species like Arabidopsis or yeast to add value to new genomes, like rice which is now available at MIPS in MOsDB. Sputnik integrates genomes for which only EST data is available (see talk by Stephen Rudd).

Return to Previous Page or Intl-PAG Homepage