Hi Eric,
1) I'm not sure about all of the mystery "tables" you found, but as for
the one example you gave, agilentCgh -- that's the name of a composite
track. So, it's not a table, or a view. You can identify these in
trackDb by the line that says "compsiteTrack on", a couple lines under
the track name . Then after the composite track entry, you'll see
indented entries for the "tracks" that make up the composite track.
Those entries (e.g., agilentCgh1x1m below) do correspond to a table --
so for example, you'll find a table called agilentCgh1x1m in hg19.
track agilentCgh
shortLabel Arrays
longLabel Microarray probesets
compositeTrack on
group varRep
visibility hide
noScoreFilter .
noInherit on
type bed 3 .
track agilentCgh1x1m
parent agilentCgh
shortLabel Ag CGH 1x1m
longLabel Agilent SurePrint G3 Human CGH Microarray 1x1M AMADID 021529
color 0,128,0
type bed 4 .
priority 1
If you find one that is definitely not a composite track, just let me
know, and I'll take a look.
2) Hiram commented in this one already. I'll check with him to see if he
came up with anything.
- Greg
On 2/28/11 3:36 PM, Eric J. M. Smith wrote:
> Greetings,
>
> Thanks to help from Maximilian and Brooke, I now have a Savant plugin
> which is happily accessing the UCSC database. I do have a couple of
> questions about parts of the database layout which are still unclear
> to me:
>
> 1) There are a number of tables which are listed in hg19.trackDb which
> aren't actually tables (e.g. agilentCgh). I thought they might be
> views, but doing "SELECT * FROM hg19.agilentCgh" from the mysql
> command prompt tells me "Table 'hg19.agilentCgh' doesn't exist". I
> would have thought that if agilentCgh were a view, the SELECT * would
> just work? Or am I missing something here?
>
> 2) A number of tables store their info in an external file (e.g.
> phastCons46way). I recognise that
> "/gbdb/hg19/multiz46way/phastCons46way.wib" is a path to a WIB file on
> UCSC's FTP server. Is there a way to access these files via HTTP, or
> is FTP the only way? Accessing byte-ranges via FTP isn't fun.
>
> Thanks again,
>
> Eric Smith
> University of Toronto
> _______________________________________________
> Genome maillist - [email protected]
> https://lists.soe.ucsc.edu/mailman/listinfo/genome
_______________________________________________
Genome maillist - [email protected]
https://lists.soe.ucsc.edu/mailman/listinfo/genome
