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] Feedback

Trish Whetzel whetzel at stanford.edu
Fri Mar 18 01:15:17 PDT 2011


Hi Stephan,

The discussion is underway here. I hope to have more details to send tomorrow.

Trish


On Mar 17, 2011, at 8:39 AM, Stephan Schurer PhD wrote:

> Sam, thank you. That's what I thought. It's still not clear to me why
> specifically "percent viability" is shown incorrectly, we have other
> concepts that have these constructs.
> Let's see what Trish and the engineering folks suggest. We do not want
> to downgrade BAO to OWL 1.0, but we need to find a way to visualize it
> correctly in NCBO.
> Need the experts' help for this.
> Trish, should I contact somebody specifically or is this already in progress?
> 
> Thanks,
> 
> Stephan
> 
> 
> 
> On Thu, Mar 17, 2011 at 2:56 AM, Saminda Abeyruwan <samindaa at gmail.com> wrote:
>> Protege 3.X is based on OWL 1.0 spec and Protege 4.X is based on OWL 2.0
>> spec and suppors 1.0 spec as well. We can express some of BAO concepts
>> according to OWL 1.0. In places where we use nominals in expression (e.g.,
>> definition of IC50, 50 percent inhibition) we will get into problems. We use
>> special constructs in OWL 2.0 to express theses concepts (OWL 2.0 include
>> those constructs specifically to support those kind of modeling
>> requirements).
>> Sam
>> On Thu, Mar 17, 2011 at 12:54 AM, Stephan Schurer PhD
>> <stephan.schurer at gmail.com> wrote:
>>> 
>>> Trish, Thank you. I guess the Protege 3 vs 4 may have to do with
>>> OWL2.0 like Sam suggested.
>>> Is there anything we can do about this? Can we trim down the version
>>> to be displayed somehow.
>>> 
>>> Uma, can you send Trish a few examples of imported GO and NCBO terms
>>> (and IDs) please.
>>> 
>>> Thanks,
>>> 
>>> Stephan
>>> 
>>> On Thu, Mar 17, 2011 at 12:50 AM, Trish Whetzel <whetzel at stanford.edu>
>>> wrote:
>>>> Updates:
>>>> - I removed the Explore link from BioPortal production, leaving the BAO
>>>> file available for download
>>>> - BioPortal uses Protege 3.4.5 in the backend, and in comparing BAO in
>>>> Protege 3.4.5 versus Protege 4.1 the hierarchy errors appear to be due to
>>>> differences between these versions of Protege. I have submitted a bug on
>>>> this for the developers to investigate further
>>>> - Can you send a few examples of imported / linked external classes from
>>>> GO and NCBO that are shown by ID, and not the label?
>>>> 
>>>> Thanks,
>>>> Trish
>>>> 
>>>> 
>>>> On Mar 16, 2011, at 9:26 PM, Stephan Schurer PhD wrote:
>>>> 
>>>>> This is a question for Trish. I also checked the consistency with the
>>>>> reasoners. I guess Trish is already looking into this.
>>>>> Trish, let us know if we can check anything on our side or if there
>>>>> are debugging tools at NCBO. Also, is there some testing site to
>>>>> upload. With this we can just try a few things / versions w/o having
>>>>> to show it publicly. It is strange that just this class was messed up.
>>>>> It would be relevant to check if this behavior is reproducible or not.
>>>>> 
>>>>> Thanks,
>>>>> 
>>>>> Stephan
>>>>> 
>>>>> On Thu, Mar 17, 2011 at 12:16 AM, Saminda Abeyruwan
>>>>> <samindaa at gmail.com> wrote:
>>>>>> BAO ontology reasoned with all consistencies checked through Pellet /
>>>>>> Pellet
>>>>>> (incremental) and Hermit 1.3.x. This implies that the DL is consistent
>>>>>> with
>>>>>> the models.
>>>>>> Is there a way to get debug/loading etc information from NCBO folks.
>>>>>> Sam
>>>>>> 
>>>>>> 
>>>>>> On Wed, Mar 16, 2011 at 10:57 PM, Vempati, Uma
>>>>>> <UVempati at med.miami.edu>
>>>>>> wrote:
>>>>>>> 
>>>>>>> Stephan,
>>>>>>> I see the same problems that you pointed out. Sam can you comment if
>>>>>>> the
>>>>>>> display of BAO at NCBO BioPortal
>>>>>>> (http://bioportal.bioontology.org/visualize/45388) has anything to do
>>>>>>> with
>>>>>>> the modeling of BAO? Or is it an error generated during parsing?
>>>>>>> 
>>>>>>> Thanks,
>>>>>>> Uma
>>>>>>> ________________________________________
>>>>>>> From: Stephan Schurer PhD [stephan.schurer at gmail.com]
>>>>>>> Sent: Wednesday, March 16, 2011 5:10 PM
>>>>>>> To: Trish Whetzel
>>>>>>> Cc: support at bioontology.org; Vempati, Uma; Saminda Abeyruwan
>>>>>>> Subject: Re: [bioontology-support] Feedback
>>>>>>> 
>>>>>>> Hallo Trish - thank you. But there is an error in the visualization.
>>>>>>> If you download the ontology and open it in Protege it is correct.
>>>>>>> Specifically the class 'percent viability' must not be at the root,
>>>>>>> but under 'bioassay component' > endpoint > response.
>>>>>>> And the class bioassay in missing from the root.
>>>>>>> Another issue is that the imported / linked external classes from GO
>>>>>>> and NCBO are shown by ID, and not the label. Again in Protege I see
>>>>>>> the label, which would be much easier to users.
>>>>>>> 
>>>>>>> I can re-upload the file, but not sure if this will change anything.
>>>>>>> 
>>>>>>> I also see a root level class for all disjoint classes and
>>>>>>> individuals, but this is not wrong.
>>>>>>> 
>>>>>>> I ask Uma to look for other potential errors.
>>>>>>> 
>>>>>>> Do you know what is going on; can you please help. I include the
>>>>>>> uploaded file here, but even if you download the file from NCBO and
>>>>>>> open in protege it is shown correct.
>>>>>>> 
>>>>>>> Thanks,
>>>>>>> 
>>>>>>> Stephan
>>>>>>> 
>>>>>>> On Wed, Mar 16, 2011 at 4:42 PM, Trish Whetzel <whetzel at stanford.edu>
>>>>>>> wrote:
>>>>>>>> Sure, the owl/xml uploaded entry is removed and the new upload of
>>>>>>>> BAO
>>>>>>>> has finished parsing and can now be explored.
>>>>>>>> 
>>>>>>>> Trish
>>>>>>>> 
>>>>>>>> 
>>>>>>>> 
>>>>>>>> On Mar 16, 2011, at 11:45 AM, Stephan Schurer PhD wrote:
>>>>>>>> 
>>>>>>>>> Hallo Trish,
>>>>>>>>> 
>>>>>>>>> Just uploaded rdf/xml. It is the same ontology version, but there
>>>>>>>>> is
>>>>>>>>> now a duplicate entry for BAO:
>>>>>>>>> http://bioportal.bioontology.org/ontologies/45388
>>>>>>>>> 
>>>>>>>>> Can you remove the one where the explore link gives an error.
>>>>>>>>> 
>>>>>>>>> Thanks,
>>>>>>>>> 
>>>>>>>>> Stephan
>>>>>>>>> 
>>>>>>>>> On Wed, Mar 16, 2011 at 2:10 PM, Trish Whetzel
>>>>>>>>> <whetzel at stanford.edu>
>>>>>>>>> wrote:
>>>>>>>>>> Yes rdf/xml.
>>>>>>>>>> 
>>>>>>>>>> Trish
>>>>>>>>>> 
>>>>>>>>>> On Mar 16, 2011, at 11:01 AM, support at bioontology.org wrote:
>>>>>>>>>> 
>>>>>>>>>>> Name : Stephan Schurer
>>>>>>>>>>> 
>>>>>>>>>>> Email : sschurer at med.miami.edu
>>>>>>>>>>> 
>>>>>>>>>>> Comment:
>>>>>>>>>>> BioAssay Ontology Cannot be explored, i.e. gives an error
>>>>>>>>>>> message.
>>>>>>>>>>> I uploaded as owl / xml. Should I re-up load as rdf / xml?
>>>>>>>>>>> 
>>>>>>>>>>> Please advise.
>>>>>>>>>>> 
>>>>>>>>>>> Thanks,
>>>>>>>>>>> 
>>>>>>>>>>> Stephan
>>>>>>>>>>> _______________________________________________
>>>>>>>>>>> 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