Linking Evolution to Genomics Using Phenotype Ontologies

From phenoscape

This was the front page of this project wiki from June 1, 2007 to July 31, 2011, while it was funded (under the title of this page) by NSF grant BDI-0641025.

Linking Evolution to Genomics Using Phenotype Ontologies

Error creating thumbnail: Unable to save thumbnail to destination

About this project

What are the developmental and genetic bases of evolutionary differences in morphology across species? Currently it is difficult to approach this question due to a lack of computational tools that allow researchers to integrate developmental genetic and comparative morphological/anatomical data.

Error creating thumbnail: Unable to save thumbnail to destination
Error creating thumbnail: Unable to save thumbnail to destination

We are addressing this by developing a database of evolutionarily variable morphological characters for a large clade of fishes (the Ostariophysi) and connecting this database to the large collection of mutant phenotypes in the ZFIN database, the central database of the zebrafish model organism community. The evolutionary and mutant phenotypes are being described using common ontologies. The database with its web-interface, together with the extended ontologies and data curation tools, will allow researchers to ask novel questions about the genetic and developmental regulation of evolutionary morphological transitions. Tool and database development are being guided by use cases, or driving research questions, defined by the devo-evo community. These tools are being developed under an open-source, open-development model, and in such a way that they can be used for additional biological systems in the future.

Error creating thumbnail: Unable to save thumbnail to destination

This project is a unique collaboration between evolutionary and model organism biologists including two national centers (NESCent and NCBO), the ZFIN model organism database, the Cypriniformes Tree of Life project, the DeepFin Research Coordination Network, and the morphological image databases used by the evolutionary biology community.

The Role of Ontologies

Error creating thumbnail: Unable to save thumbnail to destination

Background

Ontologies are constrained, structured vocabularies with well defined relationships among terms. Ontologies represent the knowledge of a particular discipline and provide not only a mechanism for consistent annotation of data, but also greater interoperability among people and machines. The most widely used biological ontology is the Gene Ontology, which is utilized to annotate molecular function, biological processes and subcellular localization to gene products from different organisms.

Phenotype ontologies

EAV4 layers flat2.png

Approximately 500 mutant zebrafish lines (alleles) with over 660 annotated phenotypic characters from the jaw or gill arches, fins, axial skeleton and other features of the skeleton have been described. Curators in the Zebrafish Information Network (ZFIN) are annotating mutant phenotypes using the zebrafish anatomy ontology and the Phenotype And Trait Ontology (PATO). PATO is a “universal” ontology of terms describing qualities (e.g. shape, color, size) that may be applied to any organism.

Anatomical ontologies

We have initiated a multi-species ontology for ostariophysan fishes, the Teleost Anatomy Ontology (TAO) (Dahdul et al, 2010), which was initialized with the terms in the zebrafish anatomical ontology. The development of the TAO is currently focused on the skeletal system because it varies significantly across the Ostariophysi, is well-preserved in fossil specimens, and it is often the focus of morphologically-based evolutionary studies in ichthyology.

This multi-species anatomy ontology is being used in combination with the PATO ontology (see EQ format) to describe the comparative morphological characters. We have also developed a separate catalog of homology statements for entities within the TAO, so that individual investigators may select particular relationships based on evidence.

Taxonomic ontology

Together with taxonomic experts, we have developed a taxonomic ontology (based on the Catalog of Fishes) to relate species with particular characters and states. The taxonomic ontology will include nodes ancestral to the Ostariophysi as far back as the Vertebrata in order to associate certain anatomical terms with more inclusive clades than the Ostariophysi.

Fish Morphology

Although the comparative anatomy of fishes has been documented in the literature for several hundred years, it is not available in a computable format. With the help of taxon experts for ostariophysan fishes, we prioritized 76 papers for curation; these can be viewed on our publicly available Google spreadsheet. 51 have been completely curated and consistency checked and are available for searching in the Knowledgebase. We achieved our goal to input approximately 5,000 morphological features (Dahdul et al, 2010) in an “EQ” format (Mabee et al. 2007a) using a combination of ontologies.

Contact

Paula Mabee (University of South Dakota) is the Principal Investigator. Co-principal investigators are Todd Vision (University of North Carolina, Chapel Hill), Monte Westerfield (University of Oregon, ZFIN), and Hilmar Lapp (NESCent) (see their contact addresses).

Acknowledgments

This project was funded by NSF grant BDI-0641025, and supported by the National Evolutionary Synthesis Center (NESCent), NSF #EF-0423641.

This project arose from a NESCent Working Group led by Paula Mabee and Monte Westerfield, "Towards an Integrated Database for Fish Evolution." Goals and summaries of the group are archived on this wiki.
nsf_logo.jpg