Data External Data DAS DAS in EnsemblEnsembl provides a DAS reference server which gives access to genomic sequences, the latest Ensembl gene predictions and repeats. The Ensembl DAS reference server with a table of available sources may be found at:
http://servlet.sanger.ac.uk:8080/das/
Here is a properly formatted XML document with the same information:
http://servlet.sanger.ac.uk:8080/das/dsn
DAS request URLs have a specific format:
protocol://site-prefix/das/data-source/command?arguments
For example:
Ensembl supports use of `stylesheet` DAS request which provides a facility to make the display of the features nicer. The full document in PDF format on what is supported may be found at:
Ensembl Stylesheet Support
DAS clients are built into several Ensembl displays to enable incorporation of third-party annotation data by "attaching a DAS source". Ensembl uses DAS for showing external third-party data and data from in-house DAS sources. The help pages for DASConfView explain how to attach a DAS data source to an Ensembl display.
Some Ensembl displays introduce semantic extensions to the DAS specification. This is explained below.
A genomic DAS source serves data in the format specified in the DAS/1.5 specification. In Ensembl, genomic DAS sources can be attached to ContigView and CytoView (see examples in human) if the annotation is in chromosomal coordinates, on contigs or on any basic assembly structure known to Ensembl for that species. For genomic DAS sources the assembly version is important as otherwise features from a source annotation (e.g. NCBI 34 human assembly) may be offset in ContigView.
Please note that for most species, Ensembl chromosomes are named like "1", "2", "X", "Y" rather than like "chr1", "chr2", "chrX", "chrY". The exceptions are the Drosophila melanogaster, Anopheles gambiae, Caenorhabditis elegans, and Saccharomyces cerevisiae genomes which uses other naming conventions.
Proteomic DAS sources (or ProteinDAS sources) can be attached to ProtView which displays protein information (see example in human).
A proteomic DAS client and server exchange protein annotations using the DAS protocol. Genomic DAS sources annotate DNA sequence from a genomic assembly in contrast to proteomic DAS sources which annotate proteins. The extension of the DAS protocol to proteins is only semantic and the requests and replies between client and server uses the exact same specification as in the genomic case. Sections of the XML reply that are not relevant to proteomic data (e.g. "phase" and "strand"), are simply ignored by the proteomic DAS client.
In most cases, the UniProt peptide sequences are used as the common reference.
Ensembl currently provides two proof-of-concept ProteinDAS annotation servers:
There is also an official UniProt reference server. This server is updated every second to serve data from the UniProt group at the EBI and also serves peptide sequences:
Example requests would be:
The UniProt DAS reference server is able to respond to any of the types of identifiers listed on the UniProt DAS information page.
ProteinDAS annotations can be browsed from Ensembl ProtView pages, e.g.
The ProtView display of Ensembl is able to display non-positional features in the same way as GeneView (see below).
DAS sources serving non-positional features (or GeneDAS sources) can be attached to GeneView (see example for human) which displays gene information. Non-positional feature DAS is a semantic extension to the DAS protocol. It allows exchange of annotations tied to identifiers, such as a HUGO gene name or an Ensembl gene ID, rather than to a segment of reference sequence. Positional data is irrelevant and the annotation applies to the entire entity referenced by the request. GeneDAS annotations are defined as having 'start' and 'end' attributes set to 0 (zero) and the annotation itself is carried in the 'note' attribute.
Among others, Ensembl currently uses the UniProt ProteinDAS reference server (see above) in a GeneDAS capacity. An example DAS request with both GeneDAS and ProteinDAS annotations is:
The corresponding Ensembl GeneView page is:
DAS sources can be automatically attached and configured while following links to Ensembl data displays.
URLs for automatic DAS source attachment must contain the following parameters:
url - The URL to the DAS server ending with '/das'.
dsn - The DAS source name as defined in the DAS server.
type - The mapping type, which defines the coordinate sytem to use for fetching features.
Possible values are:
name - The DAS source name as it will appear on Ensembl data displays.
active - Set active=1 to not only attach the DAS source but also automatically switch it on for display.
stylesheet - Set stylesheet=y to use a source stylesheet to display features.
score - Set score=h to display data as a histogram.
label - Set label=TEXT to define a name for the DAS track in ContigView and CytoView.
caption - Set caption=TEXT to define a name of the DAS source as it will appear in the DAS menu.
group - Set group=y to group features.
color - Set color=COLOUR to use this colour for feature display.
strand - Set strand=b|f|r to display features on both, forward or reverse strand, respectively.
depth - Set depth=INT for the number of rows to be used by this DAS track in ContigView. Set to 10000 for unlimited.
labelflag - Set labelflag=n|o|u to display no feature labels at all or to display them over or under the feature, respectively.
The following example URL activates a DAS track in Ensembl ContigView, which annotates the results of Ensembl Pepetide ID to PDB protein structure mapping.
http://www.ensembl.org/Homo_sapiens/contigview?peptide=ENSP00000354398&add_das_source=(url=http://das.sanger.ac.uk/das+dsn=ensppdbmapping+type=ensembl_peptide+name=ENSP_PDB_mapping+active=1)
