A Model Organism Database for Phytozome

The current back end database for Phytozome, the Joint Genome Institute's comparative plant genomics system (www.phytozome.net) has proven useful for capturing information of the proteome annotation for organisms, the analysis of the proteome, and the grouping of annotations into families of related organisms. It is a relatively denormalized schema that also lacks an extensible framework for including genomic data (e.g, diversity data). With the increasing scale and changing nature of future data in the project we have looked to moving the data store to Chado, an extensible, fully normalised, and widely used schema. We have chosen the Chado schema for the representation of our data.

Minor modifications are needed in implementing the Chado schema in order to include multiple versioned assemblies and annotation sets of the same organism, and in order to improve loading and access performance. In addition we have developed tools for in-database analytics for specialized applications.

The transition of Phytozome to the Chado schema is ongoing but expected to be completed by the next Phytozome release (v10). At this time our Chado database contains the assemblies, annotations and analyses of 41 plants.