Open mbrush opened 6 years ago
Here is a cleaner way to write the query: http://yasgui.org/short/SJOmPMOiW
It's possible there are instances of the same restriction used in both sub- and superclasses. In this case the nonredundant graph would link only at the level of the superclass.
Nevermind. I think protege is double counting uses of a term. I think we do query them all out afterall.
Can't say at the moment why the counts differ. But i ran this past one of our editors and she said that these two properties x_biomarker_of are not being maintained because the group is working on a new approach in support of one of the offices here. We can expect it out in another month or so; I can post more details in the near future.
My query for Disease-Biomarker associations doesn’t seem to be pulling back all the records it should. There are two relations in NCIT I am querying over. In the owl file, the first is used 38 times, and the second 58 times. Spot checking their use, there don’t seem to be any that are implemented as 'role groups' or in OR statements. So I expect the query to return 96 records. But it only returns 51. Any idea what is going on @balhoff?
Link to query and results here.
As an aside - I use a rather inelegant way of returning which of the two possible relations applies for each association returned. Ideas on how to better do this?