FW: [Bioperl-l] Question about Fgenesh.pm
Jason Stajich
jason at cgt.duhs.duke.edu
Fri Jul 23 13:25:00 EDT 2004
Done - would be good to have some parsing for FGENESH as well. If someone
would volunteer that would be great.
-jason
On Fri, 23 Jul 2004, Andrew Nunberg wrote:
>
> On 7/23/04 11:58 AM, "Andrew Nunberg" <anunberg at oriongenomics.com> wrote:
> Sorry I figured it out
> The parsing of the seqid was incorrect
> line 375 has
> if(/^Seq name:\s+(\S+)/) {
> $seqname = $1;
> next;
> }
> The regex should be changed to /\s+Seq name:\s+(\S+)/
> That fixes my problem since I can get the info from seq_id rather than from
> analysis_subject
>
>
> >> Hi,
> >> I am trying to parse an fgenesh analysis file using Bio::Tools::Fgenesh from
> >> bioperl-live
> >>
> >> I had to make a change in the module at line 320
> >> from:
> >> $predobj->primary_tag($ExonTags{$flds[3]} . 'Exon');
> >>
> >> to:
> >> $predobj->primary_tag($ExonTags{$flds[4]} . 'Exon');
> >>
> >> For the parser to work.
> >>
> >> I then wish to retreive the name of the dna sequence that was fed to the
> >> parser using the analysis_subject/query method
> >>
> >> $fgenesh = Bio::Tools::Fgenesh->new(-file=>"results file");
> >> $gene = $fgenesh->next_prediction;
> >>
> >> $fgenesh->analysis_subject;
> >>
> >> This method returns undef, although it appears in the code that the info is
> >> stored there.
> >> Lines 366-373 from Fgenesh.pm
> >> if(/^(FGENESH)\s+([\d\.]+)/) {
> >> $self->analysis_method($1);
> >> $self->analysis_method_version($2);
> >> if (/\s(\S+)\sgenomic DNA/) {
> >> $self->analysis_subject($1);
> >> }
> >> next;
> >> }
> >> Any input would be appreciated
> >
> > --
> > Andrew Nunberg
> > Bioinformagician
> > Orion Genomics
> > (314)-615-6989
> > www.oriongenomics.com
> >
>
> ------ End of Forwarded Message
>
> _______________________________________________
> Bioperl-l mailing list
> Bioperl-l at portal.open-bio.org
> http://portal.open-bio.org/mailman/listinfo/bioperl-l
>
--
Jason Stajich
Duke University
jason at cgt.mc.duke.edu
More information about the Bioperl-l
mailing list