![\"\"](\"/assets/images/myExp890.png\")
RDF and SPARQL\nare two really useful Open Standards. Bioclipse-RDF is a\nplugin for Bioclipse that provide RDF functionality, among which using remote SPARQL end points.
\n\nThe MyExperiment team has set up an excellent RDF front end.\nFor example, this is my MyExperiment account in RDF. The storage gets updated\nonce every day (at this moment), but I’m sure that will become more often in the future. The SPARQL end point\nallows us to make any query against the database that their ontologies\nsupport. The above query showed up 132 workflows when I ran it today.
\n\nNow, so far I have been using Gist to share Bioclipse scripts and I wrote\nsome Bioclipse GUI elements for downloading such gists .\nTo annotate these gists, Delicious has been used, and a listing of Bioclipse scripts can be found under the\ntags bioclipse and gist.
\n\nMyExperiment also allows to share workflows, but originally only for Taverna.\nA recent change, however, made it possible to share other types of workflows too. And, MyExperiment\nitself also allows all the annotation which we may want to do.
\n\nNow, using the Bioclipse-RDF functionality, I can query the MyExperiment database and use that information\ndo to stuff. If this stuff is a Bioclipse script, then I can just download it, as the download link of a\nworkflow is part of the RDF too, as we will see.
\n\nAs we have seen in the first article of this series ,\nthe RDF manager his a method to query a remote SPARQL end point. The complexity is mostly in formulating the SPARQL (and this one\nhappens to be available as workflow on MyExperiment too:
\n\n
This is worsened by the fact that JavaScript does not have a type of multiline Strings, so the backslashes at\nthe end of the lines are JavaScript syntax and not part of the SPARQL. To simplify the SPARQL, I will show\nbelow the SPARQL only, and not the Bioclipse script wrapping as is done in the above code snippet.
\n\nListing all Taverna 2 workflows, as shown in that earlier snippet, is done with the SPARQL:
\n\n\n\nThis query asks for a ?workflow and its ?title, and the workflow ?type must be of Class ContentType as defined in the\nmebase namespace, and we want to know the ?typetitle of that content type, because we are filtering that using a\nregular expression to contain “Taverna 2”. Well, if you cannot\nfollow this, just google for SPARQL,\nand run one of those tutorials which are abundantly present on the web.
To list all tags which have likely to do with metabolomics, I can do:
\n\n\n\nAnd I can also list all workflows that are tagged like this. Because I could not get string matching to work, I used the tag’s URI instead:
\n\n\n\nI was also interested in all MyExperiment Users in Sweden, and again, a simple SPARQL tells me where they live:
\n\n\n\nVery easy to find users, such as Duncan:
\n\n\n\nOr Pierre, who has not listed where he lives:
\n\n\n\nGiven a user, it is also easy to get the workflows he owns. Again, I am using my URI instead of combining with a search\nfor my account, because the MyExperiment SPARQL end point is not particularly fast:
\n\n\n\nEarlier in this series:
\n\n", "summary": "RDF and SPARQL are two really useful Open Standards. Bioclipse-RDF is a plugin for Bioclipse that provide RDF functionality, among which using remote SPARQL end points.", "image": "https://chem-bla-ics.linkedchemistry.info/assets/images/myExp890.png", "date_published": "2009-08-21T00:00:00+00:00", "date_modified": "2025-10-26T00:00:00+00:00", "tags": ["bioclipse","rdf","foaf","myexperiment","rdf","sparql"], "authors": [ { "name": "Egon Willighagen", "url": "https://orcid.org/0000-0001-7542-0286" } ] } ] }