ABIDE / FBIRN Terms Query

[dbkeator]

Hi, Just a reminder, we need 2 SPARQL query examples, one that has been started and one that hasn't: (1) Using Python, load the abide and fbirn TTL files and use a SPARQL query across those graphs where the user maps terms between the two TTL files explicitly in the query (example started: ABIDE_FBIRN_Query.ipynb). This example is for the use case when someone receives some NIDM TTL files, reads the definitions of terms and decides a joint query across datasets is appropriate.

(2) Using the abide (https://github.com/incf-nidash/nidm/blob/master/nidm/nidm-experiment/abide_clinical/abide_terms.ttl) and fbirn (https://github.com/incf-nidash/nidm/blob/master/nidm/nidm-experiment/fbirn_clinical/fbirn_terms.ttl) terms files, create a OWL file which maps the abide/fbirn terms to a set of integrated terms (see the bottom entity in https://github.com/incf-nidash/nidm-training/blob/master/ABIDE_to_NIDM_OM_with_OWLTerms.pdf) together and then create a SPARQL query which uses the OWL file to do the query.

[satra]

@dbkeator - you may want to use something like this as well. just adapt it to your examples.

these are all based off my freesurfer models:

http://jsfiddle.net/satra/tLgv4082/ http://jsfiddle.net/satra/e5vc4tLn/ http://jsfiddle.net/satra/Ludrr8wu/ (multidimensional)

http://jsfiddle.net/satra/8ykyexkw/ (XTK + s3 - missing the nidm piece, but can be easily added).

[nicholsn]

@dbkeator, for the purposes of the workshop I am going to move examples of abid/fbirn files into the nidm training repo

[nicholsn]

ok, so PR #8 addresses the first mapping issue now by creating a general NIDM term and declaring the others as subclasses. its basically the same idea as the owl approach sans protege... do you think we need both? Can you take a look an see if the example is illustrative enough?

PS, @satra thanks for the jsfiddles, those are great to work off of!