[Bioperl-l] problem with Bio::Align::Utilities - aa_to_dna_aln

[Chris Fields]

Sofia, 

Not sure, it's hard to diagnose unless we have something more to go on.  Unfortunately I can't look at it for a bit, but I suggest filing this as a possible bug, or maybe someone else has an answer?

chris

On Jun 28, 2011, at 10:29 AM, Sofia wrote:

> Hi,
> 
> I'm using the aa_to_dna_aln function from the BioPerl module Bio::Align::Utilities.
> 
> I've used it some time ago and now i'm just re-running an old script and i get this error repeatedly:
> 
> substr outside of string at /common/software/API/bioperl-1.6.1/lib/perl5/site_perl/5.8.8//Bio/Align/Utilities.pm line 160.
> Use of uninitialized value in string eq at /common/software/API/bioperl-1.6.1/lib/perl5/site_perl/5.8.8//Bio/Align/Utilities.pm line 161.
> 
> --------------------- WARNING ---------------------
> MSG: In sequence ENSP00000372224 residue count gives end value 1161.
> Overriding value [1107] with value 1161 for Bio::LocatableSeq::end().
> ATGGGGCGCTGGGCCTGGGTCCCCAGCCCCTGGCCCCCACCGGGGCTGGGCCCCTTCCTCCTCCTCCTCCTGCTGCTGCTGCTGCTGCCACGGGGGTTCCAGCCCCAGCCTGGCGGGAACCGTACGGAGTCCCCAGAACCTAATGCCACAGCGACCCCTGCGATCCCCACTATCCTGGTGACCTCTGTGACCTCTGAGACCCCAGCAACAAGTGCTCCAGAGGCAGAGGGACCCCAAAGTGGGGGGCTCCCGCCCCCGCCCAGGGCAGTTCCCTCGAGCAGTAGCCCCCAGGCCCAAGCACTCACCGAGGAC
> ---------------------------------------------------
> 
> I had a look at the code of Utilities.pm and the line 160 is:
> 
> my $char = substr($aa_seqstr,$i + $start_offset,1);
> 
> I was wondering why is the $start_offset applied to the amino acid sequence $aa_seqstr and not to the dna sequence $nt_seqstr(line 164)?
> 
> When i remove the $start_offset from the line 160 and add it to the line 164 i don't get any errors.
> 
> I don't know if that is a problem or if the problem are the arguments sending to the function.
> 
> The arguments i'm using are the same i used before: an alignment of protein sequences and a set of dna sequences. Did something change regarding the arguments?
> 
> I would appreciate any help you could provide.
> 
> Best regards,
> Sofia Pinto
> _______________________________________________
> Bioperl-l mailing list
> Bioperl-l at lists.open-bio.org
> http://lists.open-bio.org/mailman/listinfo/bioperl-l