POC-SoyBase Webex Conference Call 6-11-10, 2010
POC-SoyBase Webex Conference Call
Date: 6-11-10, 2010 8:30am (PDT)
In attendance:
POC members: Laurel Cooper (OSU), Ramona Walls (NYBG)
SoyBase members: Rex Nelson,SoyBase USDA-ARS, Ames, Iowa
Discussion items:
1. Hands-on session of OBO-Edit to help Rex create separate namespaces in the SoyBase ontology SoyBase Ontology page
RN: I can write a quick script to add a namespace tag to all of my records in the OBO file or I can use OBO edit? If I use OBO edit, will I have to go term by term and change the default namespace or can I do it globally?
PJ (by email): It needs to be done whole scale by script.
Rex sent the obo file and we opened it in OE to try the Namespace manager tool to create the three separate namespaces:
Top level: should be "Soybean_Ontology"
New namespaces:
- soybean_structure
- soybean_trait
- soybean_whole_plant_growth_stage
See comments from 5-27-10 meeting about the renumbering of the top level term: "SOY:0000001:Soybean_Ontology" that number is currently used for germination.
Rex can post a question on the OBO-Edit working group list to see if anyone has an easier way to do this or a script they already have tried.
Send questions to: OBO-Edit Working Group <Geneontology-oboedit-working-group@lists.sourceforge.net>
2. Update on accessing the SVN repository?
Laurel has created a folder on the PO SVN repository: SoyBase SVN on PO
Rex has decided that since ontology is mission critical for SoyBase, he is going to have the files stored on a repository at the USDA. There is the possibility that we could mirror them on our site or we just download the files from there as needed.
3. Other Items from Rex: Re: Questions about the QTL annotation file:
1) What identifier should I use in col. 1 (I put SOY_QTL following the TAIR file)
PJ (by email): Its the name of the db. Soybase.
2) We make our map symbols unique identifiers and such use them in our database, that's why Cols. 2 and 3 have the same value
PJ (by email): That's fine, but in the long run it may not be a good idea. Keep it that way for now. This will be the ID used by our system to link back to you.
3) Evidence codes. All of our QTL come from the literature so they are genetically mapped entities. The GO evidence codes don't seem to address genetic mapping directly. TAIR mostly used "IC" so that is what I used. It is not exactly correct but I don't know what other thing to put. Is there a better source for evidence codes?
PJ (by email): IAGP inferred by association of genotype from phenotype
http://www.plantontology.org/docs/otherdocs/evidence_codes.html
4) Namespaces in the OBO file. See above