[Bioperl-l] Genes from MySQL database using Bio::DB::GFF

[Scott Cain]

Marco,

After stepping my script through the debugger, I am pretty sure that
this really does come down to the incompatibilities between the
Bio::DB::GFF schema and some GFF3 files.  In this case, amusingly
enough, Lincoln's efforts to make the Bio::DB::GFF mysql adaptor
compatible with GFF3 has lead to this bug, whereas I didn't do the same
for the Postgres adaptor.  Unfortunately, I can't guarantee you that if
you were to switch to Postgres that it would work because it may miss
cases that the MySQL adaptor is getting.

You could try Bio::DB::SeqFeature (loaded with bp_seqfeature_load.pl)
which was designed to work with GFF3 files.  Welcome to the bleeding
edge :-)

Scott


On Wed, 2006-08-16 at 22:20 -0700, Marco Blanchette wrote:
> Many thanks Scott,
> I will probably follow your suggestion and start using PostGres. Besidebeing a different database engine, is their any big difference between usingPostGres and MySQL?
> Many thanks for the help, I was starting to doubt my ability to code!!
> Cheers,
> 
> Marco
> 
> On 8/16/06 10:11 PM, "Scott Cain" <cain.cshl at gmail.com> wrote:
> > Hi Marco,> > Well, it works for me :-)> > I ran this script:> > #!/usr/bin/perl  -w> use strict;> > use Bio::DB::GFF;> my $db = Bio::DB::GFF->new( -adaptor => 'dbi::pg',>                               -dsn => 'dbi:Pg:dbname=flybase');> > my @feat = $db->get_feature_by_name('FBgn0025803');> > for (@feat) {>     print "$_\n" if ($_->method eq 'gene');> }> > and got one line:> > gene:.(FBgn0025803)> > The only real difference is that this in a PostgreSQL database and not> MySQL.  I used Pg since I have that installed.  I'll blow away this> database, install MySQL and see if that makes a difference (of course,> it shouldn't, but you never know...)> > Gaah!  I ran the exact same script with a mysql Bio::DB::GFF and got> this out:> > gene:.(FBgn0025803)> gene:.(FBgn0025803)> > Looks like a bug in the mysql adaptor.  I'll see if I can track it down;> in the mean time, you could switch to a real database :-)> > Scott> > > > On Wed, 2006-08-16 at 23:30 -0400, Scott Cain wrote:>> Hi Marco,>> >> I'm working on it right now--my first guess (without doing any real>> work), I'm betting on the problem being an incompatibility between the>> GFF3 file and the Bio::DB::GFF schema.>> >> Scott>> >> >> On Wed, 2006-08-16 at 19:59 -0700, Marco Blanchette wrote:>>> Dear all,>>> >>> I am desperately trying to get a list of gene coordinates from a MySQL>>> database version of the fly genome populated using the Bio::DB::GFF module.>>> I have a list of 277 id in a text file that when parsed through the>>> following script return 279 entries (2 more entries then the number of genes>>> in the starting list).>>> >>> Here is the script:>>> >>> use Bio::DB::GFF;>>> my $db = Bio::DB::GFF->new( -adaptor => 'dbi::mysql',>>>                               -dsn => 'dbi:mysql:database=dmel_43_new');>>> while (<>){>>>     chomp;>>>     my @feat = $db->get_feature_by_name($_);>>>     for my $f (@feat){>>>         if ($f->type->method eq 'gene'){>>>         print     "Name: ", $f->name,>>>                 " Strand: ", $f->strand,>>>                 " Start: ", $f->start,>>>                 " End: ", $f->end,>>>                 "\n";>>>         }>>>     }>>> }>>> >>> I totally don¹t understand where the 2 extra entries are coming from.>>> Nothing differentiate them from each other. Moreover, when I double check>>> the MySQL database, both genes are having only a single Œgene¹ entry in the>>> fdata table.>>> >>> Is there a bug in the way I am trying to fetch the individual genes or>>> something is wrong with the latest Bio::DB::GFF module from the CVS>>> repository?>>> >>> Here is a test script and it¹s output that I am using to try to tract down>>> what the problem is. Hope this could help:>>> >>> use Bio::DB::GFF;>>> my $db = Bio::DB::GFF->new( -adaptor => 'dbi::mysql',>>>                               -dsn => 'dbi:mysql:database=dmel_43_new');>>> my %dups;>>> my ($j, $i) =0;>>> while (<>){>>>     chomp;>>>     my $id = $_;>>>     my @feat = $db->get_feature_by_name($id);>>>     my $feat_size = $#feat;>>>     $j++ if $feat_size == 2;>>>     >>>     for my $f (@feat){>>>         $i++;>>>           >>>         if (exists $dups{$f->group} && $f->type->method eq 'gene'){>>>             print     "Calling >>>", $f->group,>>>                         " ID=", $i,>>>                         " from \@feat of size $feat_size",>>>                         "\n";>>>             print     "Chr: ", $f->refseq,>>>                     " Strand: ", $f->strand,>>>                     " Start: ", $f->start,>>>                     " End: ", $f->end,>>>                     "\n";>>>             print "Offending >>>", $dups{$f->group}->[0]->group,>>>                   " ID=", $dups{$f->group}->[1], "\n";>>>             print     "Chr: ", $dups{$f->group}->[0]->refseq,>>>                     " Strand: ", $dups{$f->group}->[0]->strand,>>>                     " Start: ", $dups{$f->group}->[0]->start,>>>                       " End: ", $dups{$f->group}->[0]->end;>>>             print "\n\n";>>>          } elsif ($f->type->method eq 'gene') {>>>             $dups{$f->group} = [$f, $i];>>>          }>>>     }>>> }>>> >>> print "#### there was $j \@feat with only 2 features\n";>>> >>> Output of the test script:>>> >>> $ perl test.pl hrp36_targets.txt>>> Calling >>>FBgn0025803 ID=98 from @feat of size 2>>> Chr: 3R Strand: 1 Start: 16966463 End: 17038413>>> Offending >>>FBgn0025803 ID=97>>> Chr: 3R Strand: 1 Start: 16966463 End: 17038413>>> >>> Calling >>>FBgn0025681 ID=304 from @feat of size 2>>> Chr: 2L Strand: 1 Start: 2992964 End: 2998614>>> Offending >>>FBgn0025681 ID=303>>> Chr: 2L Strand: 1 Start: 2992964 End: 2998614>>> >>> #### there was 11 @feat with only 2 features>>> >>> With the hope someone can find out the problem...>>> >>> Cheers,>>> >>> Marco>>> >>> ______________________________>>> Marco Blanchette, Ph.D.>>> >>> mblanche at uclink.berkeley.edu>>> >>> Donald C. Rio's lab>>> Department of Molecular and Cell Biology>>> 16 Barker Hall>>> University of California>>> Berkeley, CA 94720-3204>>> >>> Tel: (510) 642-1084>>> Cell: (510) 847-0996>>> Fax: (510) 642-6062>> -- >> ------------------------------------------------------------------------>> Scott Cain, Ph. D.                                   cain.cshl at gmail.com>> GMOD Coordinator (http://www.gmod.org/)                     216-392-3087>> Cold Spring Harbor Laboratory>> >> _______________________________________________>> Bioperl-l mailing list>> Bioperl-l at lists.open-bio.org>> http://lists.open-bio.org/mailman/listinfo/bioperl-l
> 
> Marco Blanchette, Ph.D.
> mblanche at berkeley.edu
> Donald C. Rio's labDepartment of Molecular and Cell Biology16 Barker HallUniversity of CaliforniaBerkeley, CA 94720-3204
> Tel: (510) 642-1084Cell: (510) 847-0996Fax: (510) 642-6062
> 
> 
> 
> 
> 
> _______________________________________________Bioperl-l mailing listBioperl-l at lists.open-bio.orghttp://lists.open-bio.org/mailman/listinfo/bioperl-l
> 
-- 
------------------------------------------------------------------------
Scott Cain, Ph. D.                                   cain.cshl at gmail.com
GMOD Coordinator (http://www.gmod.org/)                     216-392-3087
Cold Spring Harbor Laboratory

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: This is a digitally signed message part
URL: <http://lists.open-bio.org/pipermail/bioperl-l/attachments/20060817/79bcf62b/attachment.sig>