Goals

A LinkML-based data model will be created (based on existing toxicological data structures) to enable a comprehensive representation of toxicological exposures, including time course, toxicants, and concentrations; genotype and genetic manipulations; phenotypic endpoints. The LinkML schema will be developed with the community, leverage a variety of different toxicological use cases, and be applicable across organisms.  

In addition, we will improve standards to represent all data, including (but not limited to)

• the Zebrafish Phenotype ontology (ZP), to represent zebrafish phenotypes. 

• the Ontology of Biological Attributes (OBA), to represent “normal” phenotypes.

A Toxicophenotype Annotation Toolkit will be iteratively created and refined with the community. This Toolkit will allow communities to annotate their data and submit them to the Zebrafish ToxicoPhenotype Atlas. 

Tutorials and curation guidelines will be provided to ensure inter-annotator consistency. 

Two methods for data submission will be designed and documented for more general laboratory use: 

1. Submission of images and metadata annotations using a simple file transfer protocol

2. Establishment and deployment of APIs  

Images representative of morphological phenotypes from toxicology studies, along with the metadata according to the new data model and zebrafish phenotype ontology, will be collected. This will provide the necessary image documentation, instantiating the new data model in a zebrafish “Toxicophenotype Atlas”. These images will provide a visual definition and standard for examining variations of specific phenotypes by laboratories at large. The online Zebrafish Toxicophenotype Atlas will be open-source, and allow users to explore and query exposures and phenotypes of interest and display example images demonstrating the phenotypes. These will be linked to external resources where relevant. FAIR API access to the data will be provided for third-party resources to utilize.