jason.stajich at duke.edu
Mon Aug 22 08:31:45 EDT 2005
On Aug 21, 2005, at 10:04 PM, Jason Stajich wrote:
> Lincoln -
> I'm getting these warning when using the memory or berkeleydb
> Use of uninitialized value in join or string at [MYPATH]/Bio/DB/GFF/
> Adaptor/memory/feature_serializer.pm line 17.
> This the line
> return join $;, at a;
> $; is not defined in my scripts, if I defining it warnings vanish,
> but did you mean ';' or do you want to provide a particular
> customizeable separator?
Or perhaps the undef warnings are due to empty fields in @a, since
features seem to be extracted out from the db later w/o incident.
BTW - I think the 'memory' and 'berkeleydb' implementations
effectively replace Bio::SeqFeature::Collection which also used a BDB
Btree to store features/locations and make range queries, but not the
full Bio::DB::GFF & DAS APIs.
> Also I notice in the SYNOPSIS of berkeleydb you have this
> # do queries
> my $segment = $db->segment(Chromosome => '1R');
> my $subseg = $segment->subseq(5000..6000);
> my @features = $subseg->features('gene');
> Is there a reason to introduce a new subseq API since we already
> have $seq->subseq($start,$end) for Bio::PrimarySeqI? I didn't
> check to see, but I assume this is a convention you are using
> throughout Bio::DB::GFF? Hopefully start,end will work and I assume
> your usual start => end works too.
> Jason Stajich
> Duke University
> Bioperl-l mailing list
> Bioperl-l at portal.open-bio.org
More information about the Bioperl-l