[Bioperl-l] [Bioperl-guts-l] Notification: incoming/1211 (fwd)

[Jason Stajich]

Felipe - no that's not a good solution, that check is actually useful in
most contexts.  This is happening b/c the cutter (RestrictionEnzyme
module) is trying to create a zero length fragment or an empty sequence.

Grr. I hate branching sometimes - I actually fixed this bug on the main
trunk already before the release and somehow it didn't get checked into
the branch.  That's done now, but of course a little late for the
release.  You need to instead apply this patch:

Index: Bio/PrimarySeq.pm
===================================================================
RCS file: /home/repository/bioperl/bioperl-live/Bio/PrimarySeq.pm,v
retrieving revision 1.42.2.2
diff -r1.42.2.2 PrimarySeq.pm
281c281,282
<
---
>     if( ! defined $seqstr ){ $seqstr = $self->seq(); }
>     return 0 unless( defined $seqstr);


Sorry about that.

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

---------- Forwarded message ----------
Date: Thu, 13 Jun 2002 16:40:15 -0400
From: bioperl-bugs@bioperl.org
To: bioperl-guts-l@bioperl.org
Subject: [Bioperl-guts-l] Notification: incoming/1211

JitterBug notification

new message incoming/1211

Message summary for PR#1211
	From: felipe wettstein <felipe.wettstein@alumni.ethz.ch>
	Subject: Bio::Tools::Gel, uninitialized value...
	Date: Thu, 13 Jun 2002 22:33:38  0200
	0 replies 	0 followups

====> ORIGINAL MESSAGE FOLLOWS <====

>From felipe.wettstein@alumni.ethz.ch Thu Jun 13 16:40:15 2002
Received: from mx01.nexgo.de ([151.189.8.96])
	by pw600a.bioperl.org (8.12.2/8.12.2) with ESMTP id g5DKeBdo014077
	for <bioperl-bugs@bioperl.org>; Thu, 13 Jun 2002 16:40:15 -0400
Received: from maria (87-138.F.dial.o-tel-o.net [212.144.87.138])
	by mx01.nexgo.de (Postfix) with ESMTP id F33403BD55
	for <bioperl-bugs@bioperl.org>; Thu, 13 Jun 2002 22:39:57 +0200 (CEST)
Date: Thu, 13 Jun 2002 22:33:38 +0200
Mime-Version: 1.0 (Apple Message framework v482)
Content-Type: text/plain; charset=US-ASCII; format=flowed
Subject: Bio::Tools::Gel, uninitialized value...
From: felipe wettstein <felipe.wettstein@alumni.ethz.ch>
To: bioperl-bugs@bioperl.org
Content-Transfer-Encoding: 7bit
Message-Id: <D7B27697-7F0C-11D6-B2D3-000393484DF2@alumni.ethz.ch>
X-Mailer: Apple Mail (2.482)

dear all

using Bioperl 1.0.1, mac os x

i was trying out perl module Bio::Tools::Gel, typing the sample programm
written in the pod SYNOPSIS:

     #An example of a virtual restriction digest and subsequent gel run
     ...
and i found a bug in line 12: '$seq' might be renamed to '$seq1' as in
the former initialisation:

     my $d = 'AAAAAAAAAGAATTCTTTTTTTTTTTTTTGAATTCGGGGGGGGGGGGGGGGGGGG';
     my $seq1 = Bio::Seq->new(-id=>'groundhog day',-seq=>$d);
     my $EcoRI = Bio::Tools::RestrictionEnzyme->new(-NAME=>'EcoRI');
     my @cuts = $EcoRI->cut_seq($seq1);
-----------------------------------^

then the programm runs well and prints out the expected output. except
that another errormessage is added:

     Use of uninitialized value in length at
/Library/Perl/Bio/PrimarySeq.pm line 282.

what appears three times, i guess because of three sequences that are
processed. it helps when i comment the three lines from 282-284 in
/Library/Perl/Bio/PrimarySeq.pm :

     #    if((CORE::length($seqstr) > 0) && ($seqstr !~
/^[A-Za-z\-\.\*\?]+$/)) {
     #        return 0;
     #    }

but this might not be the solution, since for other modules
PrimarySeq.pm seems to work. i let PrimarySeq.pm print out $seq and
$seqstr, what revealed that for the three times the errormessage is
generated, the value for $seqstr is empty, as expected. three times it
arrives correct, and three times not. i am sorry, for having no answer
to it. i suppose the problem is in Perl/Bio/Tools/Gel.pm somwhere after
line 134.

thats all. thank you


felipe wettstein




_______________________________________________
Bioperl-guts-l mailing list
Bioperl-guts-l@bioperl.org
http://bioperl.org/mailman/listinfo/bioperl-guts-l