[Bioperl-l] Re: Bio::DB Genbank?

[Jason Stajich]

Bio::DB::GenBank (and DB::GenPept for protein)  -- Grabs sequence data (w/
annotation).  You get back a Bio::Seq object which is already parsed out
for you with our GenBank parser (see Bio::SeqIO).

Works for me currently (until NCBI changes the interface again).  An
alternative - since we have folks at EBI / EMBL who are willing to provide
a stable web interface, is to use Bio::DB::EMBL - a small percentage of
the time NCBI is adding annotation that is not in EMBL, but typically
this is never a problem if you are interested in just sequence info and
rarely a problem for retrieving published annotation for a record.

This web interface is part of the proposed OBDA (http://obda.open-bio.org)
standard that some of the OBF people have put together in response to
developing a set of API for sequence data retrieval from remote or local
data sources.

-jason

On Fri, 12 Apr 2002, Baumohl, Jason wrote:

> Hello, I have been using Lincoln Stein's Boulder I/O genbank.pm for
> several scripts I have written.  Unfortuantely NCBI keeps changing
> their set up which causes Boulder to crash.  So I would like to change
> over to a hardier program to access GenBank records. Does your Bio::DB
> Genbank grab only the sequence of the file, or does it also grab the
> annotation as an object, or does it grab the entire record as a blob.
> If it does either of the last 2 I can use it for my scripts since I
> have coded a GenBank parser previously. Thanks, Jason Baumohl
>

-- 
Jason Stajich
Duke University
jason@cgt.mc.duke.edu