This is the new server for Please let us know if you notice anything weird while it's getting broken in.


Jump to: navigation, search

The Distributed Annotation System protocol (DAS; is a system for exchanging genomic annotations across the Internet. It works similarly to the idea of sharing the URLs of web-accessible GFF files, except that it is designed to support large data sets. When a client application needs to fetch just a subset of the data, such as a small piece of a chromosomal arm, the DAS protocol allows only the relevant annotations to be retrieved, rather than the whole data set.


To take advantage of DAS functionality, you will have to install the Perl Bio::Das module. This is available from CPAN (the Comprehensive Perl Archive Network ( or from the GMOD PPM repository. Unix users can install Bio::Das with this command:

% perl -MCPAN -e 'install Bio::Das'
Windows users can use the PPM tool:
C:\Windows> ppm
ppm> install Bio::Das
ppm> quit

You may need to issue the command "rep add gmod" if PPM complains that it cannot find Bio::Das.

A Tutorial

We start out with a short tutorial that illustrates how GBrowse DAS support works.

When you installed GBrowse, the system also installed a series of tutorial files, the location of which you can find at http://localhost/gbrowse2. We will use the C. elegans core tutorial database as the basis for creating a DAS server.

To start, find the elegans_core.conf configuration file in the tutorial conf directory and copy it to your GBrowse2 configuration directory (typically /etc/gbrowse2). Edit the master /etc/gbrowse2/GBrowse.conf configuration file to load this file by adding the following to the end (if it is not there already from previous tutorials):

description = Core C. elegans annotations
path        = elegans_core.conf

To enable the DAS server, add the following somewhere within the [GENERAL] section of the elegans_core.conf file:

# DAS reference server
das mapmaster      = SELF

What this line is doing is to declare to the DAS system that our server is authoritative for the coordinates on the current C. elegans genome example. This is appropriate if you are starting out a genome for the first time. If, however, you want to annotate against an existing set of genome coordinates, you should replace SELF with the URL of the DAS reference server that serves that genome. For example release hg16 of the human genome at UCSC corresponds to DAS URL A list of reference servers for various model organisms can be found at

The next step is to go through the configured tracks and add a "das category" to each of them. DAS uses the idea of the "category" of a feature in order to filter sets of features by their purpose. Categories include:

transcription features that have to do with RNA transcription
translation features that have to do with protein translation and function
variation mutations, deletions, polymorphisms
structural contigs, clones, reads, PCR primers
repeat repetitive elements
experimental a catch-all for experimental data
miscellaneous anything that doesn't fit in one fo the other categories

Find the [Genes] stanza and modify it to to have a das category of "transcription" as shown here:

feature      = gene
glyph        = gene
height       = 8
bgcolor      = blue
description  = 1
das category = transcription
key          = Protein-coding genes

Similarly, modify the [Alignments] track to have a das category of "similarity." You do not need to add a category to the DNA track, as it is treated specially by das. You're all done! Be sure to save the configuration file before you try the next step.

Using a web browser fetch the URL http://localhost/cgi-bin/gb2/das/dsn. This will return an XML document giving information about each of the data sources that you have configured.

<?xml version="1.0" standalone="yes"?>
     <DESCRIPTION>Core C. elegans annotations; Genes track; Protein-coding genes</DESCRIPTION>
     <DESCRIPTION>elegans_core; Alignments track; ESTs</DESCRIPTION>

This is showing that there is one configured DAS source, the "elegans_core" data set. Next test that the DAS "types" request is working. This request returns all the feature types that the database knows about. Using a web browser fetch the URL http://localhost/cgi-bin/gb2/das/elegans_core%7CGenes/types. This should return another short document confirming that the "gene" feature type is available.

The final test that the DAS server is performing correctly is to browse to the elegans_core database and to turn off all the tracks except for DNA/GC content. This should give you an empty details panel. Open up the "Upload and Share Tracks" tab, press "Import a Track URL", and type in http://localhost/cgi-bin/gb2/das/elegans_core%7CGenes and press "Update URLs." The page should now reload and display the gene models. Do this again with http://localhost/cgi-bin/gb2/das/elegans_core%7CAlignments in order to view the alignments. These tracks are not coming directly from the local database, but from the database via the DAS protocol.