[Bioperl-l] Database Retrieval

[Sendu Bala]

Sean Davis wrote:
>
> On 8/7/06 11:51 AM, "Sendu Bala" <bix at sendu.me.uk> wrote:
>
>> Actually, would there be any interest in a bioperl interface to the UCSC
>> tables? It could probably be done using their DAS server, and end up as
>> a very easy-to-use alternative to the Ensembl API for limited or one-off
>> queries.
>
> I haven't used their DAS server in a couple of years--it is probably worth
> my trying it again at some point.  I have gone the route of maintaining a
> local ucsc mirror of the database.  There are obvious advantages to doing
> this, including speed of access, no access limits, and the power and
> flexibility of SQL.  With the simplicity of MySQL, one can even rsync
> directory from UCSC's mysql data directory directly, eliminating the need to
> do a table import step.  I haven't gone so far as to implement a perl-based
> wrapper, but one could envision doing so for the most commonly-used tables,
> perhaps using Rose::DB::Object or DBIx::Class as a base with other methods
> added as needed based on the data contained in the table.

Yes, using MySQL would probably be better. Any interface would ideally 
use the UCSC MySQL server by default (with enforced X second waits), 
with an option to set the address to your own server.

Do you want to go ahead and look into making those classes for accessing 
the common tables? It's in my plan to make various aspects of genomic 
data retrieval a strength of bioperl as opposed to a surprising missing 
link (http://www.bioperl.org/wiki/Getting_Genomic_Sequences); I'll get 
to that in a few weeks but if you lay the ground work or better yet 
complete everything before then that would be great! :)


> If I get a chance, I could probably extract the parts from the UCSC website
> necessary to generate such a database-only mirror and put them on the wiki,
> but the information is all on the UCSC website under the "mirroring
> instructions" heading.

That would no doubt be useful.


Cheers,
Sendu.