This is the 'official' guide when we just need part sequences with annotations if we know the coordinates. And it seems it can work, but it turns out not to be.

   # don't want the entire sequence or more options
    my $gb = Bio::DB::GenBank->new(-format     => 'Fasta',
                                   -seq_start  => 100,
                                   -seq_stop   => 200,
                                   -strand     => 1,
                                   -complexity => 4);
    my $seqi = $gb->get_Stream_by_query($query);

My test codes are almost copied from the tutorial with a small change.

use strict;
use warnings;
use Bio::DB::GenBank;
my $gb = Bio::DB::GenBank->new (-format => 'genbank', 
                                -seq_start => 1, 
                                 -seq_stop => 2000, 
                                -strand =>1,);
my $seq_obj = $gb->get_Seq_by_acc('NM_000344');
print $seq_obj; #just for test

The error reports:

*MSG: acc NM_000344 does not exist
STACK: Error::throw
STACK: Bio::Root::Root::throw /usr/local/share/perl/5.10.1/Bio/Root/Root.pm:368
STACK: Bio::DB::WebDBSeqI::get_Seq_by_acc     /usr/local/share/perl/5.10.1/Bio/DB/WebDBSeqI.pm:195
STACK: test_gene_bank_with_sublocation.pl:26*

As NM_000344 does exists (http://www.ncbi.nlm.nih.gov/nuccore/NM_000344.3), thus I don't know how could this happen. Not only this gene accession number, but also others, even the genome assembly accession number. I've tried GI (get_Seq_by_gi), gene ID (get_Seq_by_id), none of them works. Is this a bug? Or probably did I miss something important?

I've tried both BioPerl (1.6.9) on Ubuntu11.10 with Perl 5.12.4, and BioPerl (1.6.1) on Ubuntu10.04 LTS with Perl 5.10.1. Neither of them works. Look forward to your reply. Thank you.