[Bioperl-l] possible filehandle out of scope bug between Bio::AlignIO and Bio::Tools::Run::Hmmer

[Jason Stajich]

The errors I get seem to be with the sequence file not being filled 
prior to running as I am getting FATAL errors from hmmalign - but I'm 
really not sure why this would happen.

FATAL: Failed to read any sequences from file /tmp/yHY9rwASfV/SyE1fpOHIM
Can't call method "consensus_string" on an undefined value at 
/home/jason/scott_hmmrun.pl line 14.

I don't have time to really debug - but you are probably right that a 
cleanup method is getting called, I am just not sure how.

--
Jason Stajich
jason.stajich at duke.edu
http://www.duke.edu/~jes12/

On Apr 18, 2005, at 4:59 PM, Scott Markel wrote:

> Jason,
>
> I'm pretty sure I was using -q.  I specifically add that
> option at the beginning of the hmmalign elsif block in
> Bio::Tool::Run::Hmmer's _run subroutine.  While I was
> debugging, the contents of the filehandle were just the
> Stockholm formatted alignment file - no hmmalign header.
>
> Scott
>
> Scott
>
> Jason Stajich wrote:
>
>> i think it has more to do with not passing -q into hmmalign which 
>> causes AlignIO to barf since the first few lines are header from 
>> hmmalign.
>> I checked in your changes with some tweaks.  It seems to work for me 
>> -- added a test and all.
>> -jason
>> -- 
>> Jason Stajich
>> jason.stajich at duke.edu
>> http://www.duke.edu/~jes12/
>> On Apr 18, 2005, at 3:17 PM, Scott Markel wrote:
>>> Note: The context for this message assumes the code change
>>> to Bio::Tools::Run::Hmmer that I just sent to the mailing
>>> list.
>>>
>>> I'm running BioPerl-1.4 on both Windows XP (Perl 5.8.0) and
>>> cygwin (Perl 5.8.5).  I get the same behavior when I use
>>> BioPerl-1.5.
>>>
>>> When I run the following code, I get the error message
>>>
>>>   Can't call method "consensus_string" on an undefined value
>>>   at runHmmAlign.pl line 14.
>>>
>>> If I change $factory to $::factory, so that it doesn't go out
>>> of scope when the subroutine is done, then everything is fine.
>>>
>>> My Perl debugging skills aren't what they should be, so I'm
>>> not sure how to verify the following, but it looks like the
>>> destructor for Bio::Tools::Run::Hmmer clobbers the filehandle
>>> in Bio::AlignIO.  Similar code involving Bio::Tools::Run::Hmmer
>>> and Bio::SearchIO (for hmmsearch) does not have this problem.
>>>
>>> I checked the bug list, but didn't find anything for AlignIO
>>> and filehandle.
>>>
>>> Scott
>>>
>>> ==============================
>>> use strict;
>>> use warnings;
>>>
>>> use Bio::Tools::Run::Hmmer;
>>>
>>> my $hmmFile = shift;
>>> my $sequenceFile = shift;
>>>
>>> my $in = Bio::SeqIO->new(-file => $sequenceFile , -format => 
>>> "fasta");
>>> my $sequence = $in->next_seq();
>>>
>>> my $hmmResults = runHmmAlign($hmmFile, $sequence);
>>> my $alignment = $hmmResults->next_aln();
>>> my $consensusString = $alignment->consensus_string();
>>> print("$consensusString\n");
>>>
>>> sub runHmmAlign
>>> {
>>>     my ($hmmFile, $sequence) = @_;
>>>
>>>     my $hmmResults;
>>>
>>>     eval
>>>     {
>>>         my $factory = Bio::Tools::Run::Hmmer->new("program" => 
>>> "hmmalign",
>>>                                                   "hmm"     => 
>>> $hmmFile);
>>>         $hmmResults = $factory->run($sequence);
>>>     };
>>>
>>>     if ($@)
>>>     {
>>>         die("hmmalign failed: $@\n");
>>>     }
>>>
>>>     return $hmmResults;
>>> }
>>> ==============================
>>>
>>> -- 
>>> Scott Markel, Ph.D.
>>> Principal Bioinformatics Architect  email:  smarkel at scitegic.com
>>> SciTegic Inc.                       mobile: +1 858 205 3653
>>> 9665 Chesapeake Drive, Suite 401    voice:  +1 858 279 8800, ext. 253
>>> San Diego, CA 92123                 fax:    +1 858 279 8804
>>> USA                                 web:    http://www.scitegic.com
>>>
>>> _______________________________________________
>>> Bioperl-l mailing list
>>> Bioperl-l at portal.open-bio.org
>>> http://portal.open-bio.org/mailman/listinfo/bioperl-l
>>>
>
> -- 
> Scott Markel, Ph.D.
> Principal Bioinformatics Architect  email:  smarkel at scitegic.com
> SciTegic Inc.                       mobile: +1 858 205 3653
> 9665 Chesapeake Drive, Suite 401    voice:  +1 858 279 8800, ext. 253
> San Diego, CA 92123                 fax:    +1 858 279 8804
> USA                                 web:    http://www.scitegic.com
>