Hello Ying Wang

To limit by species, add some more tables to the filter.

1) start as you have described
2) using the filter query, add in the related table gbCdnaInfo
3) more related tables will come up, add in the related table organism
4) filter using the organism table by setting
                     name does match "Homo sapiens" (include the quotes 
in the box)
5) output format must be "selected fields from primary and related 
tables" and you must include all of the data points that are being 
joined to link the tables. Start by adding in gbCdnaInfo and then 
organism. You can omit much of gbCdnaInfo (the index ids that link to 
other sequence desc. tables). For organism, you can just include the 
first two columns.

Help explaining #5 above:
http://genome.ucsc.edu/goldenPath/help/hgTablesHelp.html#FilterMultiple

Please let us know if you have any problems. I tested using dog as the 
base genome - with the default position as a filter. Some other genomes 
have very large/deep xeno tables. Limiting by genomic region is highly 
recommended.

Jennifer Jackson
UCSC Genome Bioinformatics Group


[email protected] wrote:
> Hi,
> I'm trying to use the table browser to download gtf files. I chose 'mRNA
> and EST tracks' as group, and 'other mRNAs' as track. It gives me all the
> mRNAs in the region. But I only need human mRNAs corresponding to my
> selected region of my species. Is there any trick to do this directly? Or
> if I want to parse the file myself, is there any tag I can use for human
> mRNAs?  For example, the note column of the gtf file looks like this:
> gene_id "AK051651"; transcript_id "AK051651"; How can I tell the species
> from this gene id? Thank you very much!
>
> Best,
>
> Ying Wang
> University of Illinois Urbana-Champaign
>
> _______________________________________________
> Genome maillist  -  [email protected]
> http://www.soe.ucsc.edu/mailman/listinfo/genome
>   
_______________________________________________
Genome maillist  -  [email protected]
http://www.soe.ucsc.edu/mailman/listinfo/genome

Reply via email to