=Paper=
{{Paper
|id=Vol-1795/paper33
|storemode=property
|title=AgroLD Indexing Tools with Ontological Annotations
|pdfUrl=https://ceur-ws.org/Vol-1795/paper33.pdf
|volume=Vol-1795
|authors=Stella Zevio,Nordine El Hassouni,Manuel Ruiz,Pierre Larmande
|dblpUrl=https://dblp.org/rec/conf/swat4ls/ZevioHRL16
}}
==AgroLD Indexing Tools with Ontological Annotations==
https://ceur-ws.org/Vol-1795/paper33.pdf
AgroLD indexing tools with ontological
annotations
Stella Zevio1,2 , Nordine El Hassouni2,3 , Manuel Ruiz2,3 and Pierre Larmande2,3
1
Université de Montpellier, Montpellier, France,
2
Institut de Biologie Computationelle, Montpellier, France,
3
South Green Bioinformatics Platform, CIRAD, IRD, Montpellier, France
pierre.larmande@ird.fr
Abstract. The Agronomic Linked Data project (AgroLD) is a Semantic
Web knowledge base designed to integrate data from various publicly
available plant centric data sources. The aim of AgroLD project is to
provide a portal for bioinformaticians and domain experts to exploit
the homogenized data towards enabling to bridge the knowledge. Here
we present new tools that enable ”full text search” functionalities with
Elastic clusters and enhance data annotation with ontologies.
Keywords: Plant Molecular Biology, Linked Data, Elastic
1 Introduction
Agronomy is an overarching field constituting various research areas such as
genetics, plant molecular biology, ecology and earth science. The last several
decades has seen the successful development of high-throughput technologies
that have revolutionized and transformed agronomic research. The application
of these technologies have generated large quantities of data and resources over
the web. In most cases these sources remain autonomous and disconnected.
The Agronomic Linked Data project (AgroLD) is a Semantic Web knowledge
base designed to integrate data from various publicly available plant centric
data sources. These include Gramene, Oryzabase, TAIR and resources from
the South Green platform among many others. The conceptual framework for
the knowledge in AgroLD is based on well-established ontologies: Gene Ontol-
ogy, Plant Ontology, Plant Trait Ontology (TO) and Plant Environment On-
tology (EO). The current phase (phase one) covers information on genes, pro-
teins, ontology associations, homology predictions, metabolic pathways, plant
traits, and germplasm. Information on the integrated databases, ontologies and
identifiers can be found in the documentation page (http://www.agrold.org/
documentation.jsp).
The RDF knowledge bases are accessed via SPARQL endpoints but these end-
points are more suitable for programmatic access. This requires at the minimum
2
a moderate knowledge of SPARQL which is not usually suitable to the non-
technical users (biologists). Consequently, the Semantic Web resources not be-
ing exploited completely. Alternatively, the AgroLD website provides four entry
points to access the underlying knowledge:
1. Quick Search, a faceted search plugin made available by Virtuoso, that
allows users to search by keywords to browse the knowledge contained in
AgroLD.
2. SPARQL Query Editor, that provides an interactive environment to for-
mulate SPARQL queries. The SPARQL editor is based on YASQE and YASR
tools [5].
3. Explore Relationships visualizer, an implementation of RelFinder [1]
that allows the user explore and visualize existing relationships between en-
tities.
4. Advanced Search, a query form providing entity (e.g. gene) specific infor-
mation retrieval. The Advanced Search query form is based on the REST
API suite developed under the AgroLD project.
To further enable the Quick Search functionality by retrieving more textual
information and hiding the technical details, we developed an indexation tool
that enables to communicate easily with Elastic clusters. Thus, this tool enables
to index Json files and to manage indexes (i.e. update, delete) on Elastic clusters
without using cURL. Starting from RDF graphs in AgroLD, we used this tool
to automate the indexation task and set up Elastic clusters. Furthermore, we
developed a generic annotation tool that enables communication with NCBO
Annotator [2] to annotate JSON files with ontologies available from such por-
tals (i.e. BioPortal [4] and AgroPortal [3]). We use this tool with AgroPortal
Annotator to enrich and index the Json files with additional information from
ontological terms such as labels, synonyms, parent and child terms, etc.
References
1. P. Heim et al. 2009. RelFinder: Revealing Relationships in RDF Knowledge Bases.
In Lecture Notes in Computer Science, 5887 LNCS:18287.
2. C. Jonquet et al. 2009. The Open Biomedical Annotator, in American Medical
Informatics Association Symposium on Translational BioInformatics, AMIA-TBI09,
(San Francisco, CA, USA), pp. 5660.
3. C. Jonquet et al. 2015. AgroPortal : a proposition for ontology-based services in the
agronomic domain. IN-OVIVE’15, Jun 2015, Rennes, France.
4. N. F. Noy et al. 2009. BioPortal: ontologies and integrated data resources at the
click of a mouse, Nucleic Acids Research, vol. 37, pp. 170173.
5. L. Rietvelda, and R. Hoekstraa. 2015. The YASGUI Family of SPARQL Clients.
Semantic Web Journal 0: 110.