Managing Phenotyping Data in the Integrated Breeding Platform

The Integrated Breeding Platform (IBP) of the CGIAR Generation Challenge Programme (GCP) provides information, analytical tools and related services to plant breeders to design and conduct modern breeding programs. The IBP is accessed through the Integrated Breeding Portal and is being developed in collaboration with 14 initial ‘user cases’ – breeding projects for eight crops in 32 developing countries. This ensures IBP development is driven by breeder needs.

A key feature of the Integrated Breeding Workflow System being developed as the informatics platform of the IBP is the Data Management Component of the Integrated Breeding Database (IBDB-DMS) which supports the management and querying of phenotyping data for plant breeding projects. DMS uses an Ontology Management System (OMS) to annotate all data variables with a cross product of ontology terms defining the trait (or more generally, the property) represented by the variable, the method or protocol by which it is assessed, and the scale or units in which it is expressed.

Phenotyping data is managed and stored as distinct datasets within studies so that data is easily retrieved for analysis and interpretation, and the analysis results are stored back in the DMS as new datasets. The data values themselves are stored in a sparse matrix structure with rows indexed by observation units and columns by standard variables. Standard variables are unique combinations of trait, scale and method from the OMS for which data has been collected.