From: Kai R. <kr...@Ce...> - 2006-01-11 12:29:08
|
Hello Randy, I have two suggestions for the mzData ontology. My first suggestion is th= e use of "instance_of". I think it would be more practical to define, for example, "Channeltron" as follows: [Term] id: PSI:1000107 name: Channeltron def: ... instance_of: PSI:1000026 ! DetectorType I'm not saying that "is_a=10" is wrong in this and other cases, but considering this snippet of XML (addmittedly a bit silly): <detector> <cvParam cvLabel=3D"psi" accession=3D"PSI:1000026" name=3D"mzOntology" value=3D"DetectorType"/> </detector> a program would validate this a correct use of the ontology (unless additional "intelligence" is programmed into it). This leads me to another problem: <source> <cvParam cvLabel=3D"psi" accession=3D"PSI:1000026" name=3D"mzOntology" value=3D"DetectorType"/> </source> is also quite difficult to spot as incorrect, as the ontology doesn't define where the terms can be used. The use of "part_of" could resolve this: [Term] id: PSI:1000026 name: DetectorType def: ... part_of: PSI:... ! Detector Regards Kai Randy Julian wrote: > There is a new PSI Ontology File (for use with mzData) on the docstore: > > http://psidev.sourceforge.net/docstore/view.php?sess=3D0&parent=3D7&exp= and=3D1&order=3Dname&sortname=3DASC&id=3D53&action=3Dfile_details > > There have been no new terms added (there are 204 entries). But there is now a definition section for every term, and the MS group is > currently editing definitions. Have a look and send any comments you h= ave. > > The file can be viewed and edited in the beta version of the OBO-Edit program: > > http://sourceforge.net/project/showfiles.php?group_id=3D36855&package_i= d=3D148487 > > or you can use a text editor... > > The next step is to reconcile the ontology with the second draft of the IUPAC nomenclature (2006 draft): > > http://www.msterms.com/wiki/index.php?title=3DIUPAC_Task_Group_MS_Terms= _Second_Draft > > There are terms which are needed by mzData and analysisXML which are no= t on this list, so any other instrument or data processing parameters which are desired by the PSI membership can be send to Randy > (rkj...@in...) or any other PSI-MS group member. > > The goal of the PSI-MS group is to update this file frequently, with th= e possibility of the human readable text changing to correct spelling or to adjust to IUPAC norms, but the accession number assigned to the concept is fixed and will not be changed, removed or reused. > > Thanks, > Randy Julian > > > > > ------------------------------------------------------- > This SF.net email is sponsored by: Splunk Inc. Do you grep through log files > for problems? Stop! Download the new AJAX search engine that makes searching your log files as easy as surfing the web. DOWNLOAD SPLUNK! http://ads.osdn.com/?ad_id=3D7637&alloc_id=3D16865&op=3Dclick > _______________________________________________ > Psidev-ms-dev mailing list > Psi...@li... > https://lists.sourceforge.net/lists/listinfo/psidev-ms-dev > > -- Real cats don't need names. But they often get called them. "Yaargeroffoutofityarbastard" does nicely. Terry Pratchett - The Unadulterated Cat **************************************************************** * email: kr...@ce... * * Universit=E4t Bielefeld * CeBiTec * 33594 Bielefeld * Germany **************************************************************** |