FlyBase has launched a project to improve the reporting of phenotypes via the controlled vocabulary (CV) statements to allow for greater versatility and accuracy as well as formalized curation of experimental conditions.
At FlyBase the phenotype CV line consists of: 1) a (starting) genotype and 2) a phenotype ontology term. For a genetic interaction, these two are accompanied by 3) an interaction CV term and 4) an interacting genotype.
At present the phenotype and genetic interaction statements do not distinguish between functionally distinct components (e.g. driver x driven transgene) of the reported genotypes – all are treated equally by the database pipeline and so each statement is copied onto the report page of each of the genotypic component. This leads to pollution of report pages for commonly used drivers with thousands of annotations, drowning away those where it’s the driver itself that’s the functionally most important and non-interchangeable component (e.g. activation of a particular group of neurons). FlyBase will therefore introduce a new category of ‘accessory alleles’. For genotypic components marked as ‘accessory’ during curation the annotation will by default not be displayed on their report pages.
The second major change to phenotype annotations at FlyBase is the addition of experimental conditions to the current CV line structure. Initially, the experimental conditions formally captured this way will be limited to chemical treatments and nutritional challenges, but they will be extended in future to include all manner of environmental perturbations. The basic unit of the chemical curation will consist of a ZECO (Zebrafish Experimental Conditions Ontology, developed and maintained by ZFIN) and a ChEBI (Chemical Entities of Biological Interest) ontology term. This combination of terms representing the experimental environment will be inserted into the current annotation structure alongside or instead of the genotype.