[Bioperl-l] taxonomy db flatfile: get taxon from gi?

Anna Friedlander anna.fr at gmail.com
Thu Nov 10 20:07:57 EST 2011


thanks all for the fast responses.

I'll try the bio-lite modules shalabh mentioned

On Fri, Nov 11, 2011 at 10:15 AM, Jason Stajich <jason.stajich at gmail.com> wrote:
> Here's another variant of one I wrote which is for my own purposes, the code
> at the beginning uses a NOSQL solution to storing all the ACC -> GI
> and then a second db to store GI -> TAXONID
> This is the case where I have a file of accession numbers and I want to add
> to the description line the taxonomy string.
> https://github.com/hyphaltip/mobedac-fungi/blob/master/scripts/taxonomy_lookupmissing.pl
> That's the first 165 lines, and then lookups are basically what you see on
> line 195.
> Would be good to rewrite that script below to use TokyoCabinent
> or KyotoCabinent (is newer implementation, not sure if it is faster?).
> one thing that this does is take up a lot of disk space ,but you can have
> tradeoffs between than and which compression scheme you use, which will
> impact performance of loading.
> Jason
> On Nov 10, 2011, at 12:51 PM, Bernd Web wrote:
>
> Hi Anna,
>
> Jason changed his example script from using hashes to using SQLite:
> bp_classify_hits_kingdom - classify BLAST hits by taxonomic kingdom
>
> See
> https://github.com/bioperl/bioperl-live/blob/master/scripts/taxa/bp_classify_hits_kingdom.pl
>
> It's an example script that shows how to do the tax to gi mapping for
> blast reports.
>
>
> Bernd
>
> On Thu, Nov 10, 2011 at 9:01 PM, Anna Friedlander <anna.fr at gmail.com> wrote:
>
> Hi all
>
> Does anyone know if there is a way to get a Taxonomy node and/or
>
> taxonid from a gi number using the flatfile with taxonomy db?
>
> I have blast output that I want to append taxonomic information to. I
>
> have hundreds of thousands of items to do this for, so it's not
>
> practical to use entrez to query the NCBI database.
>
> I have the GI->taxid file from the taxonomy ftp but it's 3.2GB so I
>
> think much too large to put into a hash!
>
> This was also discussed in 2009:
>
> http://bioperl.org/pipermail/bioperl-l/2009-April/029751.html but I
>
> don't think there was a conclusion?
>
> Thanks for your help
>
> Anna Friedlander
>
> _______________________________________________
>
> Bioperl-l mailing list
>
> Bioperl-l at lists.open-bio.org
>
> http://lists.open-bio.org/mailman/listinfo/bioperl-l
>
>
> _______________________________________________
> Bioperl-l mailing list
> Bioperl-l at lists.open-bio.org
> http://lists.open-bio.org/mailman/listinfo/bioperl-l
>
>



More information about the Bioperl-l mailing list