[annotation] Question about Isoform Annotation

[Karen Christie]

HI Kimberly,

Hmmm, I really don't think this is a case for IGI, it's really not a 
genetic interaction. I think this is the same class of assay that we 
discussed for the distinction between IMP and IDA. For the IMP vs IDA 
discussion we agreed that assays where the researcher is using the cell 
system as an assay for function and not comparing wt vs mutant versions 
should be IDA. I think the same logic applies here.

And to go back to a question from your original post, the with column is 
not allowed for IDA and will be filtered out by the checking script.

Especially if you're already going to annotate both genes to the CC
term for 'receptor complex', then I think making the MF 'contributes
to: GABA receptor activity' sounds fine.

-Karen



On Fri, 24 Aug 2007, Doug howe wrote:

> Though it is a direct assay, you can only detect it's role in the GABA 
> receptor activity when you introduce both subunit b and subunit c, so it 
> still seems like an IGI to me, though I could see how you might go down 
> either road.  Something like this perhaps:
>
> Subunit b:
>  'GABA receptor activity'   IDA   <--b has activity by itself
>  'GABA receptor activity'   IGI    with gene ID for subunit c   <--b has 
> altered activity when c is present
>
> Subunit c:
> contributes_to 'GABA receptor activity'   IGI   with gene ID for subunit b 
> <--c doesn't have activity by itself
> 'receptor complex'   IGI  with gene ID for subunit b
>
> -Doug
>
>
> Kimberly Van Auken wrote:
>> Thanks, Doug.
>> 
>> I had thought about that, but neither of the gene products is mutant and so 
>> I didn't
>> know if IGI was appropriate.
>> 
>> Perhaps Subunit C could get a contributes_to qualifier for an IDA 
>> annotation and
>> then I could annotate each of the subunits to the CC term receptor complex 
>> to fulfill
>> the contributes_to requirement that:
>>
>>    All gene products annotated using contributes_to must also be annotated 
>> to a cellular
>>    component term representing the complex that possesses the activity.
>> 
>> 
>> --Kimberly
>> 
>> 
>> Doug howe wrote:
>> 
>>> Sounds like a possible IDA function annotation on Subunit B and a 
>>> 'contributes_to' function annotation by IGI on subunit C??
>>> -Doug
>>> 
>>> 
>>> Kimberly Van Auken wrote:
>>> 
>>>> Hi--
>>>> 
>>>> I'm trying to decide the best way to annotate information about the C. 
>>>> elegans GABA receptor.
>>>> 
>>>> In C. elegans, functional GABA receptors are formed from one or more 
>>>> protein subunits encoded
>>>> by the same genomic locus, unc-49.
>>>> 
>>>> To test functionality, the authors of the paper expressed subunits 
>>>> separately and then together
>>>> and measured the dose-response to, and response to continuous application 
>>>> of, GABA.   One subunit
>>>> has receptor activity on its own, one does not, and expression of both 
>>>> results in a response distinct
>>>> from that of the first on its own.  To summarize:
>>>>
>>>>    Subunit B:  responds
>>>>
>>>>    Subunit C: doesn't respond
>>>>
>>>>    Subunits B and C: respond with distinct profile
>>>> 
>>>> 
>>>> If this was just one isoform, I would annotate the protein to the 
>>>> appropriate receptor activity terms
>>>> using the IDA evidence code.  But is there any way to annotate this 
>>>> complexity using IDA?
>>>> 
>>>> I can certainly annotate to the B isoform, but I don't want to make the 
>>>> same annotation to the C isoform.
>>>> 
>>>> One way I can think to capture the complexity would be to annotate to 
>>>> both B and C, but include the
>>>> other isoform in the WITH column for one B annotation and the only C 
>>>> annotation.
>>>> 
>>>> Is the WITH column completely off-limits for IDA annotations?
>>>> 
>>>> Any thoughts on this would be much appreciated!
>>>> 
>>>> Thanks,
>>>> Kimberly
>>>> 
>>>> 
>>>> 
>>>> 
>>> 
>> 
>