<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta http-equiv="content-type" content="text/html; charset=ISO-8859-1">
</head>
<body bgcolor="#ffffff" text="#000000">
<meta http-equiv="CONTENT-TYPE" content="text/html; charset=ISO-8859-1">
<title></title>
<meta name="GENERATOR" content="OpenOffice.org 3.2 (Win32)">
<style type="text/css">
<!--
@page { margin: 2cm }
P { margin-bottom: 0.21cm }
-->
</style>
<p style="margin-bottom: 0cm;" lang="en-US">Hello GMOD support
team,<br>
<br>
First of all I will present myself:<br>
<br>
My name is
Aurélien Bernard, I'm a software engineer in a french laboratory
called INRA (National Institute for Agricultural Research)<br>
I
currently work in the development team of a wheat annotation pipeline
that generate GFF files (TriAnnot pipeline:
<a class="moz-txt-link-freetext" href="http://urgi.versailles.inra.fr/projects/TriAnnot/index.php">http://urgi.versailles.inra.fr/projects/TriAnnot/index.php</a>).<br>
<br>
We
have decided to use Gbrowse (1.70) to allow our users to visualize
their results.<br>
Please note that Gbrowse is installed on a server
of the INRA bioinformatic platform (URGI) and used by several INRA
laboratories. <br>
Therefore we can't decide which version to use or
when to update Gbrowse.<br>
<br>
Since the beginning of my work with
Gbrowse I have created a custom configuration file, written some Perl
functions to exploit popup balloons (on hover and on click) and
developed a simple plugin for our proper use.<br>
On URGI's advice we
have chosen the following adaptor:<br>
<br>
db_adaptor =
Bio::DB::SeqFeature::Store<br>
<br>
However, when they used a MySQL
database we have decided to load GFF files into memory:<br>
<br>
db_args = -adaptor memory<br>
-dir
'/home/projects/triannot/GFF/'<br>
<br>
This choice can be
explained by the fact that our pipeline stores GFF result files into
a particular directory where our users can download them (or Gbrowse
can read them) for a short period of time (Old files are
automatically deleted after one or two weeks).<br>
<br>
Everything goes
well on the different GFF files I have tested separately.<br>
But, if
I try to put several GFF files (each of these files contains a
sequence in fasta format in its bottom part) in the same directory
then everything goes wrong:<br>
<br>
- My apache error file becomes
full of error messages<br>
- It becomes impossible to access to the
sequence of each supernumerary GFF file<br>
<br>
Example of error
message:<br>
<br>
[Mon May 03 17:02:43 2010] [error] [client 127.0.0.1]
print() on closed filehandle GEN3 at<br>
[Mon May 03 17:02:43 2010]
[error] [client 127.0.0.1]
\t/usr/local/lib/perl5/site_perl/5.8.5/Bio/DB/SeqFeature/Store/memory.pm
line
598, <GEN5> line 888 (#1)<br>
<br>
This problems seems to be
directly related to our choice of db_args so I give a look to the
correspondant BioPerl module (memory.pm).<br>
<br>
Line 598 is included
in a function with a strange comment line:<br>
<br>
# this is ugly<br>
sub
_insert_sequence {<br>
my $self = shift;<br>
my
($seqid,$seq,$offset) = @_;<br>
my $dna_fh =
$self->private_fasta_file or return;<br>
if ($offset == 0) { #
start of the sequence<br>
print $dna_fh ">$seqid\n";<br>
}<br>
print $dna_fh $seq,"\n";<br>
}<br>
<br>
Does a Gbrowse
user already encounter this problem ?<br>
<br>
Do you have a solution
to make it work better ?<br>
<br>
<br>
I thank you in advance for your
help !!<br>
<br>
Best regards,<br>
<br>
Aurélien<br>
</p>
<pre class="moz-signature" cols="72">--
Aurélien Bernard
Software engineer
INRA (Institut National de la Recherche Agronomique)
UMR INRA-UBP 1095
Génétique, Diversité et Ecophysiologie des Céréales
234, Avenue du Brézet
63100 Clermont-Ferrand
FRANCE
Tel: +33 (0)4.73.62.43.37</pre>
</body>
</html>