[go] error in gene_ontology_edit.obo

[Chris Mungall]

OBO-Edit detects disjointness violations - ie when a term is both a  
function and a process. Technically this isn't such a violation, as  
both parents are processes. It's a namespace assignment problem. If  
there isn't already a check for this can someone add it to the tracker?

We intend to move to a system with a true editors/deployment version  
split, this will help prevent some of these issues leaking into the  
publicly consumed version.

On Nov 15, 2007, at 1:58 PM, David Hill wrote:

> Hi Gavin,
>
> Thanks for the heads-up. We will fix this error this afternoon when  
> we are editing.
>
> David
>
> Gavin Sherlock wrote:
>> Hi all,
>>
>> I have discovered what I presume is an error in the current  
>> gene_ontology_edit.obo, cvs version: $Revision: 5.551, that I just  
>> downloaded from the GO site.
>>
>> The following entry:
>>
>> [Term]
>> id: GO:0008086
>> name: light-activated voltage-gated calcium channel activity
>> namespace: biological_process
>> def: "Catalysis of the transmembrane transfer of a calcium ion by  
>> a voltage-gated channel that is activated in response to  
>> light." [GOC:mtg_transport]
>> synonym: "light-activated voltage gated calcium channel activity"  
>> EXACT []
>> synonym: "light-activated voltage-dependent calcium channel  
>> activity" EXACT []
>> is_a: GO:0005245 ! voltage-gated calcium channel activity
>> is_a: GO:0010461 ! light-activated channel activity
>>
>> is listed as a biological process, yet its parents are molecular  
>> functions:
>>
>> id: GO:0005245
>> name: voltage-gated calcium channel activity
>> namespace: molecular_function
>>
>> id: GO:0010461
>> name: light-activated channel activity
>> namespace: molecular_function
>>
>> I think that's the only error, as fixing it in my local copy of  
>> the file fixes the error generated by my parser.  Can it be fixed,  
>> and should OBO-Edit detect this type of error automatically?
>>
>> Cheers,
>> Gavin
>>
>>
>>
>>
>
>
>