Search Mailing List Archives


Limit search to: Subject & Body Subject Author
Sort by: Reverse Sort
Limit to: All This Week Last Week This Month Last Month
Select Date Range     through    

[bioontology-support] FW: Fwd: Problem with Bioportal display

Natasha Noy noy at stanford.edu
Thu Feb 5 11:11:08 PST 2009


I think it would make sense to replace the protege-owl.jar with this  
one.

Natasha


On Feb 2, 2009, at 6:19 PM, Trish Whetzel wrote:

> Would it be worthwhile to use this jar for BioPortal?
>
> Trish
>
>
>> -----Original Message-----
>> From: Timothy Redmond [mailto:tredmond at stanford.edu]
>> Sent: Monday, February 02, 2009 4:27 PM
>> To: Alan Ruttenberg
>> Cc: Melanie Courtot; bioontology-support at mailman.stanford.edu; Trish
>> Whetzel
>> Subject: Re: [bioontology-support] Fwd: Problem with Bioportal  
>> display
>>
>>
>>> So during imports, there is one label that is picked up from iao.owl
>>> and a second (which is exactly the same) picked up by
>>> ontology-metadata.owl. I have removed the ones in iao.owl.
>>
>> Yes - your description is pretty much right on.  Protege is ignoring
>> some of the owl:Ontology declarations which it should be using to
>> determine the names of the imported ontologies.  After that imported
>> data started to get lost.
>>
>> I have debugged this and Tania fixed it in the latest svn.  It is a  
>> pure
>> and simple Protege bug and it made a mess of your imported  
>> assertions.
>> I think that you can just drop and use the attached protege-owl.  For
>> this style of ontology this is an important fix.
>>
>> -Timothy
>>
>>
>> ----- Original Message -----
>> From: "Alan Ruttenberg" <alanruttenberg at gmail.com>
>> To: "Trish Whetzel" <whetzel at stanford.edu>
>> Cc: "Melanie Courtot" <mcourtot at gmail.com>, bioontology-
>> support at mailman.stanford.edu, "Timothy Redmond" <tredmond at stanford.edu 
>> >
>> Sent: Monday, February 2, 2009 3:15:34 PM GMT -08:00 US/Canada  
>> Pacific
>> Subject: Re: [bioontology-support] Fwd: Problem with Bioportal  
>> display
>>
>> Ok. I see what is happening. The labels are defined in
>>
>> http://purl.obofoundry.org/obo/iao/2009-01-23/iao.owl and in
>> http://purl.obofoundry.org/obo/iao/2009-01-23/ontology-metadata.owl
>>
>> iao.owl is imported by obi.owl
>>
>> So during imports, there is one label that is picked up from  
>> iao.owl and
>> a second (which is exactly the same) picked up by ontology- 
>> metadata.owl.
>> I have removed the ones in iao.owl.
>>
>> However this behavior really should be fixed in protege. (Tim?)
>>
>> There may be other duplications in the OBI.owl file proper -  
>> Melanie can
>> answer that.
>>
>> -Alan
>>
>> On Mon, Feb 2, 2009 at 3:05 PM, Trish Whetzel <whetzel at stanford.edu>
>> wrote:
>>> Hi Alan,
>>>
>>> Since the import processing is based on Protégé itself that question
>>> would be best for Tim or Tania.
>>>
>>> Trish
>>>
>>>
>>>> -----Original Message-----
>>>> From: Alan Ruttenberg [mailto:alanruttenberg at gmail.com]
>>>> Sent: Monday, February 02, 2009 2:43 PM
>>>> To: Trish Whetzel
>>>> Cc: Melanie Courtot; bioontology-support at mailman.stanford.edu
>>>> Subject: Re: [bioontology-support] Fwd: Problem with Bioportal  
>>>> display
>>>>
>>>> Hi Trish,
>>>>
>>>> I'm doing some further investigation. Some of the duplication is  
>>>> being
>>>> generated via imports. Is there a description of the imports
>>>> processing that bioportal is doing that I could get?
>>>>
>>>> -Alan
>>>>
>>>> On Mon, Feb 2, 2009 at 1:33 PM, Trish Whetzel  
>>>> <whetzel at stanford.edu>
>>>> wrote:
>>>>> Yes, that was clear. Thanks.
>>>>>
>>>>> On Feb 2, 2009, at 1:14 PM, Alan Ruttenberg
>>>>> <alanruttenberg at gmail.com>
>>>>> wrote:
>>>>>
>>>>>> Yes, I will. I was pointing out that the current behavior (with  
>>>>>> the
>>>>>> file you don't want) is incorrect. I wasn't sure whether that was
>>>>>> clear. Is it?
>>>>>> -Alan
>>>>>>
>>>>>> On Mon, Feb 2, 2009 at 12:40 PM, Trish Whetzel
>>>>>> <whetzel at stanford.edu>
>>>>>> wrote:
>>>>>>>
>>>>>>> I thought that you were going to upload a file that did not have
>>>>>>> the replicated information?
>>>>>>>
>>>>>>> Trish
>>>>>>>
>>>>>>>
>>>>>>>> -----Original Message-----
>>>>>>>> From: Alan Ruttenberg [mailto:alanruttenberg at gmail.com]
>>>>>>>> Sent: Monday, February 02, 2009 12:03 PM
>>>>>>>> To: Trish Whetzel
>>>>>>>> Cc: Melanie Courtot; bioontology-support at mailman.stanford.edu
>>>>>>>> Subject: Re: [bioontology-support] Fwd: Problem with Bioportal
>>>> display
>>>>>>>>
>>>>>>>> It shouldn't write "definition" more than once no matter how  
>>>>>>>> many
>>>> times
>>>>>>>> one has
>>>>>>>>
>>>>>>>> OBI_XXXXX rdfs:label "definition".
>>>>>>>>
>>>>>>>> If I write
>>>>>>>>
>>>>>>>>
>>>>>>>> OBI_XXXXX rdfs:label "definition".
>>>>>>>> OBI_XXXXX rdfs:label "definition".
>>>>>>>> OBI_XXXXX rdfs:label "definition".
>>>>>>>> OBI_XXXXX rdfs:label "definition".
>>>>>>>>
>>>>>>>> There is still really only one triple.
>>>>>>>> -Alan
>>>>>>>>
>>>>>>>>
>>>>>>>> On Mon, Feb 2, 2009 at 12:01 PM, Trish Whetzel
>>>> <whetzel at stanford.edu>
>>>>>>>> wrote:
>>>>>>>>>
>>>>>>>>> Hmm, buggy in what way? Is {definition, defintion} still
>>>>>>>>> displayed
>>>>>>>>
>>>>>>>> with the
>>>>>>>>>
>>>>>>>>> added annotation files removed?
>>>>>>>>>
>>>>>>>>> Trish
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>> -----Original Message-----
>>>>>>>>>> From: Alan Ruttenberg [mailto:alanruttenberg at gmail.com]
>>>>>>>>>> Sent: Monday, February 02, 2009 11:59 AM
>>>>>>>>>> To: Trish Whetzel
>>>>>>>>>> Cc: Melanie Courtot; bioontology-support at mailman.stanford.edu
>>>>>>>>>> Subject: Re: [bioontology-support] Fwd: Problem with  
>>>>>>>>>> Bioportal
>>>> display
>>>>>>>>>>
>>>>>>>>>> We can, however the behavior is still buggy in P3. An RDF  
>>>>>>>>>> graph
>>>> is a
>>>>>>>>>> set of triples, not a multi-set.
>>>>>>>>>> Of course the behavior of P4 is *also* buggy. Sigh...
>>>>>>>>>> -Alan
>>>>>>>>>>
>>>>>>>>>> http://en.wikipedia.org/wiki/Multiset
>>>>>>>>>>
>>>>>>>>>> On Mon, Feb 2, 2009 at 11:35 AM, Trish Whetzel
>>>> <whetzel at stanford.edu>
>>>>>>>>>> wrote:
>>>>>>>>>>>
>>>>>>>>>>> Hi Alan and Melanie,
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> Can a new OBI file be uploaded that does not have the
>>>> definitions
>>>>>>>>>>
>>>>>>>>>> added to
>>>>>>>>>>>
>>>>>>>>>>> all the files, ie the change that Alan made for p4?  
>>>>>>>>>>> BioPortal
>>>> uses
>>>>>>>>>>
>>>>>>>>>> Protégé
>>>>>>>>>>>
>>>>>>>>>>> 3.4.
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> Thanks,
>>>>>>>>>>>
>>>>>>>>>>> Trish
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> From: bioontology-support-bounces at lists.stanford.edu
>>>>>>>>>>> [mailto:bioontology-support-bounces at lists.stanford.edu] On
>>>> Behalf Of
>>>>>>>>>>
>>>>>>>>>> Trish
>>>>>>>>>>>
>>>>>>>>>>> Whetzel
>>>>>>>>>>> Sent: Saturday, January 31, 2009 9:56 AM
>>>>>>>>>>> To: bioontology-support at lists.stanford.edu
>>>>>>>>>>> Subject: [bioontology-support] Fwd: Problem with Bioportal
>>>> display
>>>>>>>>>>> Begin forwarded message:
>>>>>>>>>>>
>>>>>>>>>>> From: Alan Ruttenberg <alanruttenberg at gmail.com>
>>>>>>>>>>> Date: January 31, 2009 8:56:12 AM PST
>>>>>>>>>>> To: Trish Whetzel <plwhetzel at gmail.com>, Benjamin Dai
>>>>>>>>>>> <benjamin.dai at stanford.edu>
>>>>>>>>>>> Cc: Melanie Courtot <mcourtot at gmail.com>
>>>>>>>>>>> Subject: Problem with Bioportal display
>>>>>>>>>>>
>>>>>>>>>>> http://bioportal.bioontology.org/visualize/39300/IAO_0000018
>>>>>>>>>>>
>>>>>>>>>>> Note that labels are repeated.
>>>>>>>>>>> Regardless of whether the label statement is asserted in  
>>>>>>>>>>> more
>>>> than
>>>>>>>>
>>>>>>>> one
>>>>>>>>>>>
>>>>>>>>>>> file, identical triples are considered one, and so duplicate
>>>> labels
>>>>>>>>>>> should never appear. In all cases, in a logical expression,
>>>>>>>>>>> only
>>>> a
>>>>>>>>>>> single label should be shown. Should there be more than  
>>>>>>>>>>> one, a
>>>>>>>>
>>>>>>>> tooltip
>>>>>>>>>>>
>>>>>>>>>>> can supply further information.
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> I recall that annotation properties were added in
>>>>>>>>
>>>>>>>> duplicate/triplicate
>>>>>>>>>>
>>>>>>>>>> in
>>>>>>>>>>>
>>>>>>>>>>> this release of OBI. In the email to the obi-devel list  
>>>>>>>>>>> titled
>>>>>>>>>>
>>>>>>>>>> Rendering
>>>>>>>>>>>
>>>>>>>>>>> Problem:
>>>>>>>>>>>
>>>>>>>>>>> Caused by me. Possible lesser of two evils. p4 doesn't  
>>>>>>>>>>> like it
>>>> if
>>>>>>>>
>>>>>>>> the
>>>>>>>>>>>
>>>>>>>>>>> annotation properties aren't defined in the file that uses
>> them.
>>>> So
>>>>>>>>
>>>>>>>> I
>>>>>>>>>>>
>>>>>>>>>>> added definitions to all the files. I've been debugging
>>>> performance
>>>>>>>>>>> and this was done while doing so. I can get rid of them, but
>>>>>>>>>>> if
>>>> it's
>>>>>>>>>>> not too much trouble I'd prefer to leave them. It's a bug  
>>>>>>>>>>> that
>>>>>>>>
>>>>>>>> protege
>>>>>>>>>>>
>>>>>>>>>>> displays the same thing more than once though.
>>>>>>>>>>> -Alan
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> As follow-up I have reviewed OBI in both Protégé 3.4 build  
>>>>>>>>>>> 519
>>>> and
>>>>>>>>>>
>>>>>>>>>> Protégé
>>>>>>>>>>>
>>>>>>>>>>> 4. In Protégé 4, the labels are not repeated, however in
>>>>>>>>>>> Protégé
>>>> 3.4
>>>>>>>>>>
>>>>>>>>>> they
>>>>>>>>>>>
>>>>>>>>>>> are, ie {definition, definition}. I have reported this as a
>>>>>>>>>>> bug
>>>> with
>>>>>>>>>>> BioPortal (Bug #1043). Alan, did you report this on the
>>>>>>>>>>> Protégé
>>>>>>>>>>
>>>>>>>>>> mailing
>>>>>>>>>>>
>>>>>>>>>>> list?
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>
>>>>>>>
>>>>>
>>>
>>>
> <protege-owl.jar>_______________________________________________
> bioontology-support mailing list
> bioontology-support at lists.stanford.edu
> https://mailman.stanford.edu/mailman/listinfo/bioontology-support




More information about the bioontology-support mailing list