W3C home > Mailing lists > Public > public-semweb-lifesci@w3.org > March 2007

Re: updated bams instance style model

From: Alan Ruttenberg <alanruttenberg@gmail.com>
Date: Thu, 15 Mar 2007 17:01:04 -0400
Message-Id: <27FCF956-3B41-46F3-826A-D70EDA1FF743@gmail.com>
Cc: John Barkley <jbarkley@nist.gov>, public-semweb-lifesci@w3.org
To: Luis Marenco <luis.marenco@yale.edu>

Hi Luis,

I think I understand what you are doing with the properties. You have  
(in abstract syntax)

Class(Subiculum_ventral_part_stratum_radiatum partial)
ObjectProperty(Subiculum_ventral_part_stratum_radiatum-has
	       range(unionOf(alpha2_subunit_nicotinic_receptor
			      sulfhydryl_oxidase
			      alpha4-2_subunit_nicotinic_receptor
			      alpha4-1_subunit_nicotinic_receptor))
	       domain(Subiculum_ventral_part_stratum_radiatum))

What this says, is that if you have X  
Subiculum_ventral_part_stratum_radiatum-has Y, then X must be a  
Subiculum_ventral_part_stratum_radiatum
and Y can be either a alpha2_subunit_nicotinic_receptor or a  
alpha4-2_subunit_nicotinic_receptor or a ...

A more common idiom would be:

ObjectProperty(has domain(Brain_Parts) range(Brain_Parts))  // is  
this right? or is the range protein/receptor?

The following says for any X where   
Subiculum_ventral_part_stratum_radiatum has X, X must be
either a alpha2_subunit_nicotinic_receptor, or a sulfhydryl_oxidase  
or a....
It doesn't say is actually has any Xs, though (neither does the the  
current version)

Class(Subiculum_ventral_part_stratum_radiatum
       restriction(has allValuesFrom
		      unionOf(alpha2_subunit_nicotinic_receptor
			      sulfhydryl_oxidase
			      alpha4-2_subunit_nicotinic_receptor
			      alpha4-1_subunit_nicotinic_receptor)))

The following says that every Subiculum_ventral_part_stratum_radiatum  
has at least one
alpha2_subunit_nicotinic_receptor and at least one sulfhydryl_oxidase  
and at least one ...

Class(Subiculum_ventral_part_stratum_radiatum partial
       intersectionOf(restriction(has someValuesFrom 
(alpha2_subunit_nicotinic_receptor))
		     restriction(has someValuesFrom(sulfhydryl_oxidase))
		     restriction(has someValuesFrom 
(alpha4-2_subunit_nicotinic_receptor))
		     restriction(has someValuesFrom 
(alpha4-1_subunit_nicotinic_receptor))))

Do either of these capture what the "has" relationship is supposed to  
mean? Is the object of a "has" relationship always a some sort of  
protein/receptor?

-Alan



On Mar 15, 2007, at 4:26 PM, Alan Ruttenberg wrote:

>
> Hi Luis,
>
> Thanks for this! I'm starting to have a look...
>
> In the following, owl:range should be owl:Class, I think.
>
> <owl:ObjectProperty rdf:ID="Subiculum_ventral_part_stratum_radiatum- 
> has">
>   <rdfs:domain  
> rdf:resource="#Subiculum_ventral_part_stratum_radiatum"/>
>    <rdfs:range>
>      <owl:Class>
>        <owl:unionOf rdf:parseType="Collection">
>         <owl:range rdf:about="#alpha2_subunit_nicotinic_receptor"/>
>         <owl:range rdf:about="#sulfhydryl_oxidase"/>
>         <owl:range rdf:about="#alpha4-2_subunit_nicotinic_receptor"/>
>         <owl:range rdf:about="#alpha4-1_subunit_nicotinic_receptor"/>
>      </owl:unionOf>
>     </owl:Class>
>   </rdfs:range>
> </owl:ObjectProperty>
>
> The class named "_" isn't defined - was this intended?
>
> <owl:Class rdf:about="#BAMS-Atlases">
>   <rdfs:subClassOf rdf:resource="#_" />
> </owl:Class>
>
> Also, BAMS-Atlases is class without subclasses or instances - was  
> that your intention?
>
> What's the idea behind making all the object properties, rather  
> than, for instance a single has, has-target, etc?
>
> -Alan
>
>
> On Mar 15, 2007, at 2:50 PM, Luis Marenco wrote:
>
>>
>> Hi All,
>>
>> I helped Mihai Bota design the prototype BAMS xml protocol (for  
>> anatomical data)
>> for the purpose of facilitating export of its contents into  
>> Ontology servers.
>> (e.g.: QIS system. In very alpha version: http://os- 
>> qis.med.yale.edu ), There
>> some vocabularies and simple ontologies can be automatically  
>> imported, mapped,
>> and allowed to be exported in several experimental protocols.
>>
>>
>> For Bams, an Owl dump can be accessed at
>> http://os-qis.med.yale.edu/site/tools/export/owl_export2.aspx?sid=22
>>
>> You can import this ontology directly into Protege following the  
>> next steps
>> . Select -> File\New Project
>> . Check the box "[ ]Create from Existing sources"
>> . Select Project type "OWL/RDF Files"
>> . Click [next] button
>> . Copy and paste
>> "http://os-qis.med.yale.edu/site/tools/export/owl_export2.aspx? 
>> sid=22"
>> . Click [Next >] button
>> . Select "OWL DL"
>> . Click [Next >] button
>> . Select "Logic View"
>> . Click [Finish] button
>>
>>
>> Use the same stpes to import an old ontology version of NeuronDB
>> http://os-qis.med.yale.edu/site/tools/export/owl_export2.aspx?sid=9
>>
>> ModelDB
>> http://os-qis.med.yale.edu/site/tools/export/owl_export2.aspx?sid=19
>>
>> and others ...
>>
>> Thanks,
>>
>> ________________________
>> Luis Marenco
>> Yale Center for Medical Informatics
>>
>>
>> -----Original Message-----
>> From: public-semweb-lifesci-request@w3.org
>> [mailto:public-semweb-lifesci-request@w3.org] On Behalf Of John  
>> Barkley
>> Sent: Thursday, March 15, 2007 2:22 PM
>> To: public-semweb-lifesci@w3.org
>> Subject: updated bams instance style model
>>
>>
>> I put the updated bams instance style model on the wiki page
>> http://esw.w3.org/topic/HCLS/HCLSIG_DemoHomePage_HCLSIG_Demo
>>
>> The new version modifies the old version according to Alan's
>> suggestions from tuesday's f2f.
>>
>> jb
>>
>>
>>
>>
>
>
Received on Thursday, 15 March 2007 21:02:04 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 14:52:30 UTC