Biodas : Can I Retrieve Some Metadata About A Segment ?

Question

[Pierre Lindenbaum]

Hi all,

I wonder if there is a biodas way to retrieve some metadata from a feature?segment= query

For example: in http://genome.ucsc.edu/cgi-bin/das/hg18/features?segment=22:14504273,14622020;type=knownGene

http://www.biodas.org/dtd/dasgff.dtd">
<DASGFF>
<GFF version="1.0" href="&lt;a href=" http:="" genome.ucsc.edu="" cgi-bin="" das="" hg18="" features"="" rel="nofollow">http://genome.ucsc.edu/cgi-bin/das/hg18/features">
<SEGMENT id="22" start="14504273" stop="14622020" version="1.00" label="22">
<FEATURE id="uc002zkr.2.chr22.14504263.0" label="uc002zkr.2">
 <TYPE id="knownGene" category="transcription" reference="no">knownGene</TYPE>
 <METHOD></METHOD>
 <START>14504264</START>
 <END>14504973</END>

 <SCORE>-</SCORE>
 <ORIENTATION>-</ORIENTATION>
 <PHASE>-</PHASE>
 <GROUP id="uc002zkr.2.chr22.14504263">
  <LINK href="&lt;a href=" http:="" genome.ucsc.edu="" cgi-bin="" hgTracks?position="chr22:14504263-14572999&amp;db=hg18" "="" rel="nofollow">http://genome.ucsc.edu/cgi-bin/hgTracks?position=chr22:14504263-14572999&db=hg18">Link to UCSC Browser</LINK>
 </GROUP>
</FEATURE>

<FEATURE id="uc002zkr.2.chr22.14504263.1" label="uc002zkr.2">
 <TYPE id="knownGene" category="transcription" reference="no">knownGene</TYPE>
 <METHOD></METHOD>
 <START>14542397</START>
 <END>14542487</END>
 <SCORE>-</SCORE>
 <ORIENTATION>-</ORIENTATION>

 (...)

What is 'uc002zkr.2' (geneSymbol, cds Start , ... ) ? I know I can retrieve this information through a mysql query but does BIODAS define a standard protocol to retrieve this kind of indormation ?

Thanks,

Pierre

Answer 1

[Jeremy Leipzig]

From what I can see UCSC does not even implement the DAS1.5 search-by-id spec, so actually getting more information about a feature this way seems highly unlikely.

none of these work, for example

http://genome.ucsc.edu/cgi-bin/das/hg18/features?feature_id=uc002zkr.2.chr22.14504263.0 http://genome.ucsc.edu/cgi-bin/das/hg18/features?group_id=uc002zkr.2.chr22.14504263.0 http://genome.ucsc.edu/cgi-bin/das/hg18/features?id=uc002zkr.2.chr22.14504263.0

[Pierre Lindenbaum]

Thanks Jeremy , I didn't know those syntaxes for the DAS URLs

http://genome.ucsc.edu/cgi-bin/das/hg18/features?id=

Answer 2

[Maximilian Haeussler]

I see that you mention the mysql server already, but I'm adding some detail nevertheless: UCSC is not supporting DAS very well. It's because they have thousands of meta data tables and it would be hard to decide how to summarize them into a single string.

It's easier to use the Table Browser's "describe schema" field to find the tables linked to your table of interest and look for the field you're interested in. Then run mysql queries against the public mysql server, as you suggested. Mysql is a lot more flexible than any DAS API could ever be. https://genome.ucsc.edu/goldenPath/help/mysql.html.