[ensembl-dev] Gene Symbol

[Nick Fankhauser]

Yes, thanks a lot! Like this it produces a lot less false hits.
Especially when I combine it with rejecting all MHC chromosomes.

But there's still for example SF3B1 and RAGE, for which I get results
from two different chromosomes for some reason. Do you know why this can
still be the case?

Nick


On 13/02/12 17:00, Andy Yates wrote:
> Hi Nick,
> 
> My guess is that you're hitting an issue with external synonyms. The method you are using will consult all xrefs linked to a gene (along with transcripts and translations) as well as consulting the external_synonym table. In the case of CLK2 we have the following external synonyms linked to the term CLK2.
> 
> synonym	db_name	dbprimary_acc	display_label
> clk2	Vega_transcript	OTTHUMT00000364143	OTTHUMT00000364143
> clk2	Vega_transcript	OTTHUMT00000365664	RP11-531A21.3-001
> clk2	Vega_transcript	OTTHUMT00000272912	OTTHUMT00000272912
> clk2	OTTG	OTTHUMG00000150164	OTTHUMG00000150164
> CLK2	EntrezGene	9894	TELO2
> clk2	HGNC	2069	CLK2
> 
> If you change your query to limit by the external DB of the source then the hits will reduce massively e.g.
> 
> my $genes = $gene_adaptor->fetch_all_by_external_name('CLK2', 'HGNC');
> 
> All the best
> 
> Andy
> 
> On 13 Feb 2012, at 15:22, Nick Fankhauser wrote:
> 
>> Hi!
>>
>> I'm trying to retrieve the chromosomal position for a list of
>> gene-symbols. They are all official gene-symbols.
>>
>> Using a loop like this
>>
>>    foreach my $gene
>> (@{$gene_adaptor->fetch_all_by_external_name($gene_symbol)}) {
>>
>> I get one correct hit for some genes (e.g. USP8), but for others like
>> CLK2, I get multiple results and have no idea how to select the correct
>> one. Is there a way to just get position of just the official gene symbol?
>>
>> Thanks!
>>
>> _______________________________________________
>> Dev mailing list    Dev at ensembl.org
>> List admin (including subscribe/unsubscribe): http://lists.ensembl.org/mailman/listinfo/dev
>> Ensembl Blog: http://www.ensembl.info/
> 
> 
> _______________________________________________
> Dev mailing list    Dev at ensembl.org
> List admin (including subscribe/unsubscribe): http://lists.ensembl.org/mailman/listinfo/dev
> Ensembl Blog: http://www.ensembl.info/