[Bioperl-l] FW: Bioperl PAML bug

[Chris Fields]

Jason,

There is a new bug for PAML results parsing:  

http://bugzilla.open-bio.org/show_bug.cgi?id=1983

I'll try looking into it but I can't get around to it until late tomorrow or
Thursday at the earliest; I'm forwarding this to you and the mail list in
case you or others might have ideas.  Looks like the results aren't parsed
past the third model result in the test results file (in the bug report).
Below is the email I sent to the bug submitter.

Cheers!

Chris

P.S.  Hope you're enjoying the break after your defense.  Congrats again!

Christopher Fields
Postdoctoral Researcher - Switzer Lab
Dept. of Biochemistry
University of Illinois Urbana-Champaign 


> -----Original Message-----
> From: Chris Fields [mailto:cjfields at uiuc.edu]
> Sent: Tuesday, April 18, 2006 8:55 AM
> To: 'nivsabath at hotmail.com'
> Subject: Bioperl PAML bug
> 
> From what I have read on the HOWTO PAML page, I believe you globbed
> together two different scripts here; in other words, the submitted script
> is NOT the same as on the web page.
> 
> =====================================
> use strict;
> use Bio::Tools::Run::Phylo::PAML::Codeml;
> use Bio::Tools::Run::Alignment::Clustalw;
> 
> # for projecting alignments from protein to R/DNA space
> use Bio::Align::Utilities qw(aa_to_dna_aln);
> 
> # for input of the sequence data
> use Bio::SeqIO;
> use Bio::AlignIO;
> ...
> =====================================
> 
> should be:
> =====================================
> use strict;
> use Bio::Tools::Phylo::PAML;
> ...
> =====================================
> 
> since you're only parsing PAML output (not running PAML and parsing the
> results).
> 
> Regardless, I still get errors here; it looks like it's not parsing past
> the first three models, so it's a legit bug.  Looks like the parser
> doesn't catch the end of the model (calling the model method time_used
> gets an "Use of uninitialized value in print" error).
> 
> Also, looks like the POD and docs for using Bioperl and PAML are a bit
> out-of-date and still refer to names of older methods (I ran into a number
> of missing method calls while looking into this).  I won't be able to get
> to this today but I'll drop Jason a line to see what he thinks.
> 
> Chris
> 
> Christopher Fields
> Postdoctoral Researcher - Switzer Lab
> Dept. of Biochemistry
> University of Illinois Urbana-Champaign