[Gramene] rice protein sequences

[Will Spooner]

On 17 Jun 2009, at 18:14, Peifen Zhang wrote:

> Hi,
>
> I am trying to retrieve rice protein sequences at Gramene, i.e.  
> LOC_Os01g01302. When I do a quick search for LOC_Os01g01302, it  
> returned two proteins, Q0JRG4 and Q655K8, both referring to  
> LOC_Os01g01302 (http://www.gramene.org/db/searches/quick_search?category=&search_for=LOC_Os01g01302&x=0&y=0 
> ).
>
> My question is why there are two proteins and two sequences for  
> LOC_Os01g01302? Is that because they are both best BLAST hits  
> meeting a certain cut-off?

Hi Peter,

Q0JRG4 and Q655K8 are both UniProtKB/TrEMBL proteins. These are  
automatically annotated from EMBL nucleotide sequences, and are fairly  
redundant. The mappings between UniProtKB/TrEMBL to rice genes are  
based on best-in-genome alignment, and it is common for multiple  
TrEMBL proteins to map to a single gene.


>
> Finally, is rice-proteins.fa.zip in ftp://ftp.gramene.org/pub/gramene/fasta/ 
>  the correct file to retrieve rice protein sequences (for japonica  
> group)?

This really depends on what you are looking for. If you want the set  
of rice proteins in Gramene that correspond to UniProtKB proteins,  
then this is the best file. If you want proteins that correspond to  
predicted cDNAs on the rice genome (LOC_ IDs), then this file may  
serve you better;

ftp://ftp.gramene.org/pub/gramene/release29/data/fasta/oryza_sativa_japonica/pep/Oryza_sativa_japonica.TIGR5.52.pep.all.fa.gz

All the best,

Will


>
> Thanks and regards,
> Peifen
> _______________________________________________
> Gramene mailing list
> Gramene at brie4.cshl.edu
> http://mail.gramene.org/mailman/listinfo/gramene