[Bioperl-l] Porblems with Bio::Tools::Run::Phylo::PAML::Baseml

[Albert Vilella]

Just to mention that the new paml4 has a "basemlg" instead of a
"baseml" binary. AFAIK, Jason fixed codeml to make it work both for
paml3.xx a paml4, but I am not sure about baseml.

Also, I think if you set runmode 0, you have to provide a tree:

#!/usr/bin/perl

use Bio::Tools::Run::Phylo::PAML::Baseml;
use Bio::AlignIO;
use Bio::TreeIO;
my $alignio = Bio::AlignIO->new(-format => 'phylip',
                                -file =>
'/home/avilella/bioperl/vanilla/bioperl-run/scripts/test.phy');
my $treeio = Bio::TreeIO->new(-format => 'newick',
                                -file =>
'/home/avilella/bioperl/vanilla/bioperl-run/scripts/test.tree');
my $aln = $alignio->next_aln;
my $tree = $treeio->next_tree;

my $bml = Bio::Tools::Run::Phylo::PAML::Baseml->new();
$bml->alignment($aln);
$bml->tree($tree);
$bml->executable("/home/avilella/9_opl/paml/paml3.14/src/baseml");
$bml->save_tempfiles(1);
my $tempdir = $bml->tempdir();


#set the runmode to zero
$bml->set_parameter('runmode', 0);

my ($rc,$parser) = $bml->run();
system "more $tempdir/baseml.ctl";

while ( my $result = $parser->next_result ) {
    my @otus = $result->get_seqs();
    my $MLmatrix = $result->get_MLmatrix();
    $DB::single=1;1;
    # 0 and 1 correspond to the 1st and 2nd entry in the @otus array
}
exit;

4 50
Homo_sapie AGUCGAGUC---GCAGAAACGCAUGAC-GACC
Pan_panisc AGUCGCGUCG--GCAGAAACGCAUGACGGACC
Gorilla_go AGUCGCGUCG--GCAGAUACGCAUCACGGAC-
Pongo_pigm AGUCGCGUCGAAGCAGA--CGCAUGACGGACC

ACAUUUU-CCUUGCAAAG
ACAUCAU-CCUUGCAAAG
ACAUCAUCCCUCGCAGAG
ACAUCAUCCCUUGCAGAG

(((Homo_sapie,Pan_panisc),Gorilla_go),Pongo_pigm);
On Nov 23, 2007 12:29 PM, Sendu Bala <bix at sendu.me.uk> wrote:
> Dave Messina wrote:
> > Daniel,
> >
> > I don't have bioperl-run or PAML installed on my system to test it myself,
> > but have you tried the latest version of bioperl-run from CVS? It looks like
> > that code has been worked on since 1.5.2 was released.
>
> Yes, I fixed it in CVS so it should at least /run/. I don't know about
> the parsing side of things, though that may also have been fixed
> recently by someone else.
>
>
> _______________________________________________
> Bioperl-l mailing list
> Bioperl-l at lists.open-bio.org
> http://lists.open-bio.org/mailman/listinfo/bioperl-l
>