Difference between revisions of "Chado GBrowse Summary View"

Revision as of 23:19, 3 August 2010

Starting in GBrowse version 2.13, administrators have the option of configuring a "summary view" (feature density) for tracks when zoomed out too far to effectively show individual features, and starting with the 0.30 release of Bio::DB::Das::Chado this functionality is supported for browsing Chado databases with GBrowse.

Prerequisites

• GBrowse 2.13
• Bio::DB::Das::Chado 0.30
• PostgreSQL 8.1 (probably--I only tested with 8.4, but I'm reasonably sure it will work with versions back to 8.1)

Configuration

To support the summary view, Chado needs an additional table the the Chado GBrowse adaptor will use to calculate the summary view. To create this table (gff_interval_stats) and populate it with data, the Bio::DB::Das::Chado distribution comes with a perl script, gmod_create_summary_statistics.pl. Executing this script, optionally with a --dbprofile flag to indicate the database to work on, will create the table if it doesn't exist, as well as a few stored procedures for populating the table, and then execute the stored procedure. This script can also be used in a cronjob to periodically recreate the data in the summary statistics data when the data in the table becomes stale due to added data.

To turn on summaries, add this line to your data source's conf file in the main section:

 show summary   = 100000

which will make all tracks go to the summary view when the region shown is greater than 100000. To turn summaries off for a given track, add

 show summary   = 0

to the track configuration.