[go] Alert: Proposal to obsolete GO:0048503 (function ontology) that impacts existing annotation

[Midori Harris]

Hi Alex,

My comment on the keyword mapping was not intended to mean that the 
'GPI-anchor' shouldn't be mapped to any GO term at all; I meant only 
that it should not be mapped to the new, correctly defined and 
applied, 'GPI anchor binding term'. Sorry if that wasn't clear.

m

On Fri, 18 Jan 2008, Alexander Diehl wrote:

> Midori,
>
> I did not and do not agree that the SwissProt mapping should disappear.  At a 
> minimum I would like to see the SwissProt keyword, 'GPI-anchor', mapped to 
> 'anchored to membrane ; GO:0031225'.  Per PMID:17361015, I believe this is an 
> entirely correct mapping.
>
> See 
> http://sourceforge.net/tracker/index.php?func=detail&aid=1873999&group_id=36855&atid=440764 
> for my extended comments and suggested replacement terms.
>
> --Alex
>
>
> Midori Harris wrote:
>> Hello,
>> 
>> The proposal has been made to obsolete
>>
>>   GPI anchor binding ; GO:0048503
>> 
>> Annotations to this term exist at present as follows:
>>
>>   DB        total    non-IEA
>>   GeneDB_Spombe    10    10
>>   GOA_chicken    25    2
>>   GOA_cow    30    4
>>   GOA_human    128    9
>>   GOA_uniprot    280    4
>>   MGI        123    8
>>   RGD        59    5
>>   ZFIN        11    0
>> 
>> This term is not used in any GO slim set maintained within the OBO flat
>> file.
>> 
>> This term is used in one external mapping:
>> spkw2go: SP_KW:KW-0336 GPI-anchor > GO:GPI anchor binding ; GO:0048503
>> 
>> Note that this mapping should be deleted (and not replaced when the new 
>> term becomes available), because the Swiss-Prot keyword is used for 
>> proteins that have been modified by the addition of a GPI anchor.
>> 
>> The reason for this proposal is this term has been widely, and incorrectly, 
>> used to annotate proteins that have GPI anchors covalently attached. A new 
>> term will be created that can be used to annotate gene products that 
>> interact non-covalently with GPI anchors.
>> 
>> The discussion has taken place on the annotation mailing list; a summary 
>> has been lodged in SourceForge:
>> https://sourceforge.net/tracker/?func=detail&atid=440764&aid=1873999&group_id=36855 
>> 
>> The two-week comment period ends on Friday, February 1, 2008.
>> 
>> *** Unless objections are received by February 1,
>>     we will assume that you agree to this change. ***
>> 
>> Midori
>
>
>