Yes, I already did something like that - downloaded feature tables from
NCBI site. Sorry again for asking questions unrelated to bioperl.
2013/9/26 Jason Stajich
> More likely you want another place to ask the question - biostars perhaps
> - http://www.biostars.org/
> if you were only searching genomes you could perhaps grab the ptt files
> from the FTP site - though you would need to lookup GI ids to get the ACC
> number for the NP_XX which is a bit convoluded and I'm not sure you can
> round trip.
> But the produce descriptions for the genomes like in bacteria are in
> .ptt files.
> On Sep 24, 2013, at 11:00 PM, Alexey Morozov
> > Dear colleagues,
> > I'm not sure if this list is the right place to ask, but I have a
> > regarding NCBI eutils. Say, I have lots of CDS's IDs like NP_769305.1
> > YP_001638012.1. "Lots" as in "Millions of them" (a metagenomic
> > course, I want to get something actually meaningful, so I decided to
> > "product" field of associated protein records. Can I spare fetching
> > complete records, which will easily be gygabytes upon gigabytes of
> > unnecessary data and get at least only "protein" feature of genpept
> > --
> > Alexey Morozov,
> > LIN SB RAS, bioinformatics group.
> > Irkutsk, Russia.
> > _______________________________________________
> > Bioperl-l mailing list
> > Bioperlemail@example.com
> > http://lists.open-bio.org/mailman/listinfo/bioperl-l
> Jason Stajich
LIN SB RAS, bioinformatics group.