Closed ValWood closed 8 years ago
could have a look at Dropbox/pombase/ontologies/phenotype/phaf_to_crap_gaf.pl ...
;-)))))
If I'm all packed will take a look soon.....
could have a look at Dropbox/pombase/ontologies/phenotype/phaf_to_crap_gaf.pl
Thanks Midori. I like the script name.
Can this issue be closed now?
I think so... we are done until we hear back from Seth
Hi Kim,
No urgency, but when you get a moment could you do a quick perl to munge the PHAF file into a GAF to try the AmiGO installation:
GAF 1 -> PHAF 1 yes (though we've actually omitted this column from the files Kim loads into Chado because we haven't taken any external files to date)
GAF 2 -> PHAF 2 yes
GAF 3 -> PHAF 10 very dodgy, because GAF 2 and 3 refer to the SAME object, whereas PHAF 10 refers to a specific allele, NOT the same db object as the gene; I would go with the gene name as we do with actual GAFs, i.e. PHAF 9
GAF 4 -> n/a yes
GAF 5 -> PHAF 3 yes
GAF 6 -> PHAF 18 yes
GAF 7 -> PHAF 13 yes
GAF 8 ->n/a yes
GAF 9 -> n/a yes
GAF 10 -> n/a true, there's no exact equivalent, but this would be a safer place than GAF 3 for PHAF 10
GAF 11 ->n/a true, there's no exact equivalent, but could use PHAF 11 (it's no worse, and maybe slightly better, than mapping GAF 3 to PHAF 10)
GAF 12 -> n/a true, there's no exact equivalent, but could use PHAF 12 (it's no worse, and maybe slightly better, than mapping GAF 3 to PHAF 10)
GAF 13 -> PHAF 19 yes
GAF 14 -> PHAF 20 yes
GAF 15 -> n/a will always be "PomBase" for the foreseeable future, so we could plug it in
GAF 16 -> 15, 16 and 17 yes
GAF 17 -> n/a no exact equivalent, but could use for an allele ID
(I was hoping I could do it with cut, but you can't change the output order, hence I'd need to do a quick script...if anyone can do before Kim gets to it, go ahead......I might but probably not this week)