[Bioperl-l] commits coming in

Jason Stajich jason at cgt.duhs.duke.edu
Fri May 16 15:36:42 EDT 2003


As soon as pub.open-bio.org is back online and I have network time I'll
make the commits of the changes I did on the train this week.  Amazing
what you can accomplish with no email and less distractions.


First off I've written Bio::Taxonomy::Node which is a simple
implementation which is database aware (through the Bio::DB::Taxonomy
interface I've started).  It is mainly written to be a companion to
Bio::DB::Taxonomy::flatfile to get taxonomy information.  A companion
script in scripts/taxa/local_taxonomydb_query.PLS shows some simple
examples - I am not sure about implementing proper tests since the initial
indexing of the TaxaDB can take a while.  This will eventually implement
all of the functionality of Bio::Species plus have the ability to walk up
and down the Taxonomy throuugh parent/child relationships.  A little bit
more work needs to be done and a Bio::DB::Taxonomy::biosql driver will
also fit this role.  The Bio::DB::Taxonomy::entrez is not quite rich
enough for our needs, but I need to play around more with the interface -
if we can just get parent IDs for a taxonomy node then we are basically
set.  The plan is eventually to be able to grab all children of a give
node so that one can, for example, get the taxids for all 'Gammaproteo
bacteria', etc.

Also, a BSML writer based on Incogens XLST of NCBI blast XML to BSML has
been written in Bio::SearchIO::Writer::BSMLResultWriter.  This is the
first phase in an attempt to play nicer with the richer formats and
eventually may one of many paths to loading data into a Chado schema in
concert with software developed at TIGR.  A companion script
search2bsml.PLS is written as well.

Hopefully this will be committed tomorrow or Sunday.

-jason

--
Jason Stajich
Duke University
jason at cgt.mc.duke.edu


More information about the Bioperl-l mailing list