This proposal came out of a presentation by Erick Antezana and Frederic Potier at the August 2009 GMOD Meeting titled:
From Frederic and Erick:
We were thinking about having CMap features displayed in GBrowse in one or more dedicated tracks. Using now the fact that GBrowse can be connected to multiple datasources, the GBrowse config file could be for example:
#################################
# database definitions
#################################
# Database for genome annotations.
[GGB_TIGR_RICE_V5:database]
db_adaptor = Bio::DB::GFF
db_args = -adaptor dbi::mysql
-dsn dbi:mysql:GGB_TIGR_RICE_V5:localhost:3306
-user ...
-pass ...
# Database for NGS
[bamtest:database]
db_adaptor = Bio::DB::Sam
db_args = -fasta sequence.fasta
-bam bam_file.bam
# Database for genetic markers
[CMAP_TIGR_RICE_V5:database]
db_adaptor = specific CMAP adaptor or maybe an already existing one ???
db_args = -adaptor dbi::Oracle
-dsn dbi:Oracle:CMAP_TIGR_RICE_V5:localhost:
-user ...
-pass ...
...
### TRACK CONFIGURATION ####
...
[Pair]
feature = read_pair
glyph = segments
database = bamtest
draw_target = 1
show_mismatch = 1
bgcolor = sub {
my $f = shift;
return $f->get_tag_values('M_UNMAPPED') ? 'red' : 'green';
}
fgcolor = green
height = 3
label = sub {shift->display_name}
label density = 50
bump = fast
connector = dashed
key = Read Pairs
[AFLP_RICE]
feature = aflp
database = CMAP_TIGR_RICE_V5
map_set = tigr_rice_v5
key = CMAP AFLP markers
category = Molecular Mapping
[SSR_RICE]
feature = aflp
database = CMAP_TIGR_RICE_V5
map_set = tigr_rice_v5
key = CMAP SSR markers.
category = Molecular Mapping
....
As you can see, both AFLP_RICE and SSR_Rice are pulled out from the CMap database and displayed in dedicated tracks.
Then the idea is to manage the links to CMap via an url in the popup ballon.
This configuration should be very convenient to integrate CMap/GBrowse without requiring a lot a maintenance and administration.