[Myexperiment-discuss] myExperiment REST API/SPARQL endpoint and other...

[Kostas Karasavvas]

Hi all and thanks for your replies!

I had a quick look at the REST API but first let me give a brief update about my current project.

I recently started working at the Netherlands Bioinformatics Centre (NBIC) and I am based in LUMC. I am working with Marco Roos on eLab components (and other things).

My current project involves enabling the execution of taverna workflows from inside galaxy (http://galaxy.psu.edu/) effectively making the whole taverna workflow run as a single tool (that is how they call the workflow tasks in galaxy).

There is a proof of concept that shows how it can be done. We now want to generalize for more taverna workflows. The idea is to create a script, that given the appropriate information (see below), it creates the galaxy tool (to be installed to the galaxy server).

Then a link could be placed in myExperiment so that users can download the workflow as a galaxy tool (that was already discussed between Marco and Don, I believe). 

To achieve this I need certain wkf information: 

1) general title

2) general description

3) inputs (source) name, description, example, type and depth, e.g. for T2 workflows

4) outputs (sink) name, description, example, type and depth

I checked the REST API and it seems I get access only for 1) and 2)   (please, correct me if I am wrong).

I checked the RDF files (downloaded one directly:  http://.../workflows/74.rdf) and I found that it contains more information than the REST API like I/Os names (title), example and type (which, seems that it will be a string most of the times). But no description or depth. Regarding the latter, how can one know if the input to the taverna wkf is a list (not sure how rare that is in practice but in theory it can happen) or a file?

Since the information about inputs and outputs has been extracted already from the workflows, how easy would it be to make them available in the REST API?

The SPARQL endpoint could also be used to access the some of the information I want from the RDF files. But I would still need this extra information: I/O descriptions and depth. I know both will be available for scufl2 workflows. I am not sure if t2flow includes them (?).

Sorry for the long email! I hope you can make some sense out of it :)

Thank you for reading,

Kostas

PS. Don/David, I added you in skype for further discussion, if need be.