Search Mailing List Archives

Limit search to: Subject & Body Subject Author
Sort by: Reverse Sort
Limit to: All This Week Last Week This Month Last Month
Select Date Range     through    

'with' field contents for IPI annotations

Evelyn Camon camon at
Wed Mar 1 06:08:10 PST 2006

Great harold...

We will also accept UniProt Isoform Ids (P00001.1, P00001.2) if you can 
figure out which is the correct one in the paper and its has an id in 
UniProt..bit of an awkward one at the moment.


Harold Drabkin wrote:
> If there are multiple isoforms known, but it is not clear from the paper 
> which was used, we use the generic UniProt id. It it is clear, we would 
> use the Uniprot with the correct suffix.  If there are no UniProts,  
> next in line would be a Trembl record, then  an NP_ if available, then 
> protein_id; if no protein ids at all, we use a nucleic acid id if it can 
> figured out.  We used to put gene identifiers (MGI:1234567) in the past, 
> but have converted then.
> If, however, it is "binds actin", we'd use "actin binding" with an IDA 
> and no seqid/gene_id at all, since there are several actin genes.
> Evelyn Camon wrote:
>> thanks however..we integrate data from Consortium member association 
>> files so how data is added there and why is very important to us...
>> ..the issue I was addressing was whether we should convert a gene id
>> to a protein Id if we can map it or if a gene id should be left as it 
>> is provided as there were specific reasons why a protein id was not 
>> provided...curator not sure of interacting protein from publication...
>> It is of course easier not to convert any ids from external sources...
>> thanks
>> Evelyn
>> Midori Harris wrote:
>>>> How will we know the difference between a group adding gene ids 
>>>> instead of protein Ids as the latter was not available and group 
>>>> adding a gene id for other reasons???
>>> For the purposes of the gene_association file, I think it doesn't 
>>> matter.
>>> Of course if a database has trouble finding UniProt IDs, they should
>>> communicate with UniProt to address the problem, but that should all
>>> happen separately (and by human communication, not file parsing) -- 
>>> i.e.,
>>> there's no point trying to capture it in the gene_association file.
>>> m
>>>> If Consortium member has difficultly getting Uniprot ids we should 
>>>> address these concerns and the appropriate contacts can be made.
>>>> I appreciate these thing can take time and solution is needed in the 
>>>> interim.

Evelyn Camon
GOA Coordinator
Senior Scientific Curator
European Bioinformatics Institute
E-mail: camon at

More information about the go-discuss mailing list