From: Helen P. <par...@eb...> - 2007-11-21 18:04:15
|
This addition of a slot makes sense to me, when we changed our data acquisition system to make users be explicit about their reference channel sample then we got usable info, before that it was just 'control' which tells us nothing, cheers Helen David Sutherland wrote: > Hiya, > > On Wed, 2007-11-21 at 17:51 +0000, Georgios V. Gkoutos (Genetics) wrote: > >> On Nov 21 2007, Alan Ruttenberg wrote: >> >> >>> [Term] >>> id: PATO:0000587 >>> name: decreased size >>> def: "A size which is lesser relative to the normal or >>> average." [PATOC:GVG] >>> subset: value_slim >>> related_synonym: "small" [] >>> related_synonym: "tiny" [] >>> is_a: PATO:0000117 ! size >>> >>> The question is: what defines the normal from which the average is >>> computed. >>> Can the definition be phrased in terms of exceed some number of >>> standard deviations from an explicitly defined population? >>> >>> Some uses in the wild: >>> http://arabidopsis.info/StockInfo?NASC_id=6262 Presumably the >>> population is some normal strain? Which one? >>> >>> In the example in the pheno representation doc (http:// >>> www.fruitfly.org/~cjm/obd/pheno-representation.html) we have >>> >>>> <phenotype> >>>> <phenotype_character> >>>> <bearer> >>>> <!-- heart --> >>>> <typeref about="MA:0000072"/> >>>> </bearer> >>>> <quality> >>>> <!-- small --> >>>> <typeref about="PATO:0000587"/> >>>> </quality> >>>> </phenotype_character> >>>> </phenotype> >>>> In PS this is represented as: >>>> >>>> E= MA:0000072 Q= PATO:0000587 >>>> >>> Is this in reference to all mouse hearts, as use to describe a >>> strain? Or with respect to a particular animal, within a litter, in >>> relation to the litter? >>> >>> Do you see the problem? >>> >> Yes,you're right. One obviously needs more information to define normality >> - information that cannot be held in PATO or MP for that matter. What level >> the description is applied to (i.e litter level, phenodeviant etc) depends >> on the observation. As you suggest, how small is defined for the MP term >> "small hear" (MP:0002188) obviously depends on the particular strain. It >> would be impractical to have all this information in MP (i.e. small heart >> for BALB/c strain or C3H etc). As PATO is meant to be applicable accross >> species this would be unachievable. So this information is meant to be held >> elsewhere. For example, a particular assay could dictate what's normal and >> how this normality is define.As the assay could be animal/strain etc. >> specific it could provide the range of normality. >> >> > I can see that this type of information couldn't (and shouldn't) be > stored in PATO. What worries me more is the standard system of > phenotype curation using terms like this from PATO. Most descriptions > of phenotypes by model organism biologists are at least implicitly > comparative (comparison to wild-type). Perhaps this could be made more > explicit in the curation system - with a slot for what the comparison is > to. This could contain extra genotypic info for the control, a strain > or simply 'wild-type' (I think there's a good argument for treating > 'wild-type' as a statement about strain). Having an extra slot would at > least push curators to think about what the comparison is to. > > David > >>> -Alan >>> >>> On Nov 21, 2007, at 11:23 AM, Alan Ruttenberg wrote: >>> >>> >>>> This worries me. Will "small" always be used as a relative quality? >>>> What will the reference be. How can we be sure that "small" will be >>>> used consistently? >>>> -Alan >>>> >>>> On Nov 21, 2007, at 7:15 AM, Bill Bug wrote: >>>> >>>> >>>>> Hi George, >>>>> >>>>> That's perfect. I'd thought I remembered seeing this before when >>>>> I browsed through PATO. Thanks, George. >>>>> >>>>> I wonder why they didn't come up when I searched in BIRNLex (which >>>>> imports all of PATO). >>>>> >>>>> Ah - two reasons: >>>>> >>>>> 1) Protege will only search the "DISPLAY SLOT" property, which I >>>>> had set to "rdfs:label". Not being able to search synonyms along >>>>> with the preferred label is definitely a problem in Protege >>>>> >>>>> 2) There may also be something about the way obo properties >>>>> translate into OWL (OBOinOWL) that is also problematic. As you >>>>> can see in the OWL/RDF/XML below, synonyms are not given simply >>>>> as string literal annotation properties such as is done with >>>>> rdfs:label, but rather they are specified as individuals of the >>>>> class "synonym" and related via a "hasRelatedSynonym" property. >>>>> >>>>> <oboInOwl:hasRelatedSynonym> >>>>> <oboInOwl:Synonym> >>>>> <rdfs:label xml:lang="en">small</rdfs:label> >>>>> </oboInOwl:Synonym> >>>>> </oboInOwl:hasRelatedSynonym> >>>>> >>>>> This adds a level of abstraction that might make it more difficult >>>>> to search for a class using synonyms, even once Protege is able to >>>>> index more than a single DISPLAY_SLOT annotation property, which I >>>>> believe is in the works. >>>>> >>>>> Thanks again for pointing me in the right direction, George. >>>>> >>>>> Cheers, >>>>> Bill >>>>> >>>>> On Nov 21, 2007, at 5:27 AM, Georgios V. Gkoutos (Genetics) wrote: >>>>> >>>>> >>>>>> Hi Bill, >>>>>> >>>>>> we have "decreased size" (PATO:0000587) and "increased >>>>>> size" (PATO:0000587) in PATO with synonyms "small" and "large". >>>>>> Would this be OK? >>>>>> >>>>>> George >>>>>> >>>>>> On Nov 21 2007, Bill Bug wrote: >>>>>> >>>>>> >>>>>>> Hi All, >>>>>>> >>>>>>> I realize this is a problematic quality to address, but in BIRN >>>>>>> we have a need to qualify aspects of specific observations as >>>>>>> "small" and "large". Now, obviously these are relative >>>>>>> measures - measures that really only can apply to continuants. >>>>>>> >>>>>>> Still - I noticed at the base of PATO there are these simple >>>>>>> qualities: >>>>>>> >>>>>>> quality >>>>>>> quantitative >>>>>>> amount >>>>>>> decreased >>>>>>> increased >>>>>>> >>>>>>> It would be helpful to also have >>>>>>> >>>>>>> quality >>>>>>> quantitative >>>>>>> size >>>>>>> small >>>>>>> large >>>>>>> >>>>>>> at least for those observations where that is all that is >>>>>>> specified. >>>>>>> >>>>>>> Has this issue come up before? >>>>>>> >>>>>>> Cheers, >>>>>>> Bill >>>>>>> >>>>>>> >>>>>>> William Bug, M.S., >>>>>>> M.Phil. email: >>>>>>> wb...@nc... >>>>>>> Ontological Engineer (Programmer Analyst III) work: (610) 457-0443 >>>>>>> Biomedical Informatics Research Network (BIRN) >>>>>>> and >>>>>>> National Center for Microscopy & Imaging Research (NCMIR) >>>>>>> Dept. of Neuroscience, School of Medicine >>>>>>> University of California, San Diego >>>>>>> 9500 Gilman Drive >>>>>>> La Jolla, CA 92093 >>>>>>> >>>>>>> Please note my email has recently changed >>>>>>> >>>>>>> >>>>>>> >>>>>>> >>>>>> -- >>>>>> George Gkoutos >>>>>> >>>>>> ---------------------------------------------------- >>>>>> Dr. G.V.Gkoutos >>>>>> >>>>>> University of Cambridge >>>>>> Department of Genetics >>>>>> Downing Site, Downing Street >>>>>> Cambridge, CB2 3EH >>>>>> >>>>>> Tel: +44 [0] 1223 333963 >>>>>> Fax: + 44 [0] 1223 333992 >>>>>> >>>>>> email (direct): gg...@ca... >>>>>> email (departmental): g.g...@ge... >>>>>> url: http://www.gen.cam.ac.uk >>>>>> >>>>>> ------------------------------------------------------ >>>>>> >>>>>> >>>>>> >>>>>> >>>>> William Bug, M.S., M.Phil. >>>>> email: wb...@nc... >>>>> Ontological Engineer (Programmer Analyst III) work: (610) 457-0443 >>>>> Biomedical Informatics Research Network (BIRN) >>>>> and >>>>> National Center for Microscopy & Imaging Research (NCMIR) >>>>> Dept. of Neuroscience, School of Medicine >>>>> University of California, San Diego >>>>> 9500 Gilman Drive >>>>> La Jolla, CA 92093 >>>>> >>>>> Please note my email has recently changed >>>>> >>>>> >>>>> --------------------------------------------------------------------- >>>>> ---- >>>>> This SF.net email is sponsored by: Microsoft >>>>> Defy all challenges. Microsoft(R) Visual Studio 2005. >>>>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ >>>>> _______________________________________________ >>>>> Obo-phenotype mailing list >>>>> Obo...@li... >>>>> https://lists.sourceforge.net/lists/listinfo/obo-phenotype >>>>> >>> ------------------------------------------------------------------------- >>> This SF.net email is sponsored by: Microsoft >>> Defy all challenges. Microsoft(R) Visual Studio 2005. >>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ >>> _______________________________________________ >>> Obo-phenotype mailing list >>> Obo...@li... >>> https://lists.sourceforge.net/lists/listinfo/obo-phenotype >>> >>> > > > ------------------------------------------------------------------------- > This SF.net email is sponsored by: Microsoft > Defy all challenges. Microsoft(R) Visual Studio 2005. > http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ > _______________________________________________ > Obo-phenotype mailing list > Obo...@li... > https://lists.sourceforge.net/lists/listinfo/obo-phenotype > -- Helen Parkinson, PhD ArrayExpress Production Coordinator, Microarray Informatics Team, EBI EBI 01223 494672 Skype: helen.parkinson.ebi |