TY - DATA T1 - Biological Pathways and the Semantic Web PY - 2012/12/04 AU - Andra Waagmeester AU - Tina Kutmon AU - Egon Willighagen AU - Alex Pico UR - https://figshare.com/articles/presentation/Biological_Pathways_and_the_Semantic_Web/101922 DO - 10.6084/m9.figshare.101922.v1 L4 - https://ndownloader.figshare.com/files/229480 KW - Semantic Web, KW - pathways KW - OpenPHACTS KW - Bioinformatics N2 - Tutorial on Biologial Pathways and the Semantic Web Andra Waagmeester (1), Martina Kutmon (1), Egon Willighagen (1), Chris Evelo (1), Alex Pico (2) (1) Department of Bioinformatics - BiGCaT, Maastricht University, Maastricht, The Netherlands(2) Gladstone Institutes, San Francisco, California, USA Exposing pathway repositories to the semantic web.At what point do you stop searching for an ontology and start developing your own?Writing SPARQL queries to extract content from a semantic pathway repository.Various ways of querying and integrating with different non-semantic platforms.Integration of pathway data with other linked data databases. BackgroundWikiPathways (www.wikipathways.org) is a community-driven pathway resource [1-3]. It is inspired by Wikipedia and aims at capturing knowledge about biological pathways in such a way that it can be used for data analysis. For the Innovative Medicine Initiative project Open PHACTS [4], we expose the contents of WikiPathways in semantic web format. We are taking a multi-pronged approach, selecting different ontologies to capture diverse aspects of semantics. As an established pathway ontology, BioPAX [5] forms the bulk of the semantics for the actual pathway content, but we also use other specialized ontologies to capture the semantics of curators, authors, and other WikiPathways-specific meta data. Data distributionOnce converted into RDF the data is available for download to be used in a variety of ways. The first is to create a SPARQL endpoint that reflects the current state of wikipathways.org. As curators make edits, the changes are converted and made almost immediately available through this SPARQL endpoint.Utilizing linked data caches, such as the one produced by Open PHACTS, the RDF content of WikiPathways can be periodically integrated with other biological and pharmacological data. We can also combine the RDF data with other pathway resources and load the entire collection into one single SPARQL endpoint. Choosing the right ontologiesCore to data conversion into a semantic form is the right choice of ontologies. For pathway knowledge BioPAX is the obvious ontology. For those concepts where there are no classes in BioPAX a different ontology is needed. The big question here is, where to find proper ontologies and when to decide to stop searching and start developing your own. In the tutorial we will survey the ontology look up services (e.g. BioPortal [6] and linked open vocabularies [7]). In our approach we decided to create two autonomous ontologies which are described at http://vocabularies.wikipathways.org. In the tutorial we will demonstrate how to map from your own custom ontology to existing ontologies such as BioPAX. Creating SPARQL queriesOnce the data is made available through a SPARQL endpoint, different queries can be extracted. We will demonstrate some pathway-relevant SPARQL queries. So called CONSTRUCT queries can be written to add triples to enrich the knowledgebase with new links to other relevant biological linked open data. For example, this can add links to other web pages by using FOAF:homepage, or use mappings of similarity with concepts in other resources to integrate information from those (using for example skos:broadMatch[8]). Using semantic data in applicationsFinally, we will address some ways of using the semantic data in platforms like Cytoscape [9] and PathVisio [10]. Both of these platforms have plugin extensions that have been developed to query SPARQL endpoints. The use of these tools will be demonstrated with examples that will show how the semantic information from pathways can be used to enrich other types of data and vice versa. As an example we will show how the information about proteins in a pathway can be extended with information about drugs able to target these proteins and thus possibly influence the pathway. AudienceThe tutorial is aimed at an audience that is interested in using biological pathway information in semantic form, either directly using our pathway information in their research or in linking it to other semantic resources or by using it as an example to use other biological pathway resources on the semantic web. If the audience includes editors of other biological knowledgebases, we could also run a small hackathon to link all of our resources. References1. Pico 2008 http://dx.doi.org/10.1371/journal.pbio.00601842. Kelder 2009 http://www.plosone.org/article/info:doi/10.1371/journal.pone.00064473. Kelder 2011 http://dx.doi.org/10.1093/nar/gkr10744. Williams 2012 http://dx.doi.org/10.1016/j.drudis.2012.05.0165. BioPortal http://bioportal.bioontology.org6. BioPAX http://www.biopax.org7. Linked Open Vocabularies http://labs.mondeca.com/dataset/lov/8. SKOS http://www.w3.org/2004/02/skos/9. Smoot 2010 http://dx.doi.org/10.1093/bioinformatics/btq67510. Van Iersel 2008 http://dx.doi.org/10.1186/1471-2105-9-399 The research leading to these results has received support from the Innovative Medicines Initiative Joint Undertaking under grant agreement n°115191, resources of which are composed of financial contribution from the European Union’s Seventh Framework Programme (FP7/2007-2013) and EFPIA companies‘ in kind contribution and the NIH National Institute for General Medical Sciences (R01-GM100039) ER -