Font Size:
The Crop Ontology: towards the semantic integration of open plant trait data
Building: Grand Hotel Mediterraneo
Room: Sala dei Continenti
Date: 2013-11-01 12:01 PM – 12:10 PM
Last modified: 2013-10-07
Abstract
In order to understand the expression of plant phenotypes arising from the interaction between genotypes and environments, scientists study and describe the developmental stages, morphological traits, functional traits and roles of plants. The lack of standardization in this nomenclature is a significant barrier to information sharing among scientists and with non-scientific communities like farmers and policymakers who have their own knowledge and representations of biodiversity. Concepts like farmers’ traits and uses of plants should be integrated in information systems focusing on in situ and on-farm conservation and use of crop diversity. Broad consensus is needed to standardize metadata to annotate and cross-link data about plant phenotypes, genotypes and environmental conditions, and to mediate between different knowledge sources.
A group of international scientists working in Plant Sciences have already articulated this vision called the Common Reference Ontology for Plants (cROP). cROP, if realized, would implement reference ontology or thesauri to support an international semantic framework for the integration of data on cultivated plants. The Crop Ontology (CO) (http://www.cropontology.org/) has been recognized as a highly valuable source of concepts to support harmonized annotation of crop trait-related data in cROP. Alongside PO, TO and the Phenotypic Quality Ontology (PATO), CO will contribute to the consolidation of the current fragmented semantic landscape of trait descriptions to facilitate queries for related gene expression and phenotype data from diverse plant databases, including various international crop germplasm collections.
The Crop Ontology project was conceived by the CGIAR Generation Challenge Programme (GCP) primarily to provide the semantics for data and tools of the GCP Integrated Breeding Platform (IBP; https://www.integratedbreeding.net/), hosted by the infrastructure of iPlantCollaborative (http://www.iplantcollaborative.org/). Therefore, the CO currently focuses on breeders’ traits, environment and experimental design variables. CO is composed of species-specific concepts that are cross-referenced online to Plant Ontology (PO) and Trait Ontology (TO) through hyperlinks. It is available in a Resource Description Framework (RDF) format served through an Application Programming Interface (API). Trait concepts can then be automatically downloaded and used by the IBP crop databases and any third party websites like the European Solanaceae breeding database, and the international cassava database.
In the future, the Crop Ontology will be further integrated through RDF schemata derived from Darwin Core extended to germplasm (https://code.google.com/p/darwincore-germplasm/) and will also support access to data captured in situ and on farm, published as Linked Open Data (LOD). LOD is a semantic web paradigm for the publication of open access machine-readable data. LOD best practices support exposing, sharing, and connecting pieces of data, information and knowledge on the Semantic Web using uniform resource identifier (URI) and RDF. A collaboration to elaborate RDF standards for agricultural LOD exchange is ongoing with the FAO OpenAGRIS project and with a working group hosted by the European-funded AgInfra project. The objective is to create crop ontology-driven web connections for the discovery of open data across research infrastructures. Considering that Biodiversity research is a multidisciplinary science, discovering data from other domains of research is key for scientists.
A group of international scientists working in Plant Sciences have already articulated this vision called the Common Reference Ontology for Plants (cROP). cROP, if realized, would implement reference ontology or thesauri to support an international semantic framework for the integration of data on cultivated plants. The Crop Ontology (CO) (http://www.cropontology.org/) has been recognized as a highly valuable source of concepts to support harmonized annotation of crop trait-related data in cROP. Alongside PO, TO and the Phenotypic Quality Ontology (PATO), CO will contribute to the consolidation of the current fragmented semantic landscape of trait descriptions to facilitate queries for related gene expression and phenotype data from diverse plant databases, including various international crop germplasm collections.
The Crop Ontology project was conceived by the CGIAR Generation Challenge Programme (GCP) primarily to provide the semantics for data and tools of the GCP Integrated Breeding Platform (IBP; https://www.integratedbreeding.net/), hosted by the infrastructure of iPlantCollaborative (http://www.iplantcollaborative.org/). Therefore, the CO currently focuses on breeders’ traits, environment and experimental design variables. CO is composed of species-specific concepts that are cross-referenced online to Plant Ontology (PO) and Trait Ontology (TO) through hyperlinks. It is available in a Resource Description Framework (RDF) format served through an Application Programming Interface (API). Trait concepts can then be automatically downloaded and used by the IBP crop databases and any third party websites like the European Solanaceae breeding database, and the international cassava database.
In the future, the Crop Ontology will be further integrated through RDF schemata derived from Darwin Core extended to germplasm (https://code.google.com/p/darwincore-germplasm/) and will also support access to data captured in situ and on farm, published as Linked Open Data (LOD). LOD is a semantic web paradigm for the publication of open access machine-readable data. LOD best practices support exposing, sharing, and connecting pieces of data, information and knowledge on the Semantic Web using uniform resource identifier (URI) and RDF. A collaboration to elaborate RDF standards for agricultural LOD exchange is ongoing with the FAO OpenAGRIS project and with a working group hosted by the European-funded AgInfra project. The objective is to create crop ontology-driven web connections for the discovery of open data across research infrastructures. Considering that Biodiversity research is a multidisciplinary science, discovering data from other domains of research is key for scientists.