[Bioperl-l] Bio::*Taxonomy* changes

[Chris Fields]

> > I don't any real objection to Bio::Taxon though if that's what you'd
> > like to name it - although, what will happen to the Bio::Taxonomy
> > hierarchy then? Phased out?
> 
> At the moment it seems to me that the Bio::Taxonomy modules (excluding
> Node) aren't really usable. Jason wrote a comment in t/TaxonTree.t which
> tests Taxon and Tree:
> 
> ## I am pretty sure this module is going the way of the dodo bird so
> ## I am not sure how much work to put into fixing the tests/module
> 
> FactoryI is strange (it isn't intended to work like any other Bioperl
> factory) and there are no implementers of it, while Taxonomy.pm itself
> would be redundant after my Node changes and has implementation issues,
> though it may make more sense to some people.
> 
> My vote is phase out.
> 
> 
> What is the actual process involved in renaming a module in Bioperl?

This is how many times the phrase "Bio::Taxonomy" is used in Bioperl in
directory Bio (which should catch any namespace usage like Node, etc.):

Instances: 2    BP Module : Bio::DB::Taxonomy
Instances: 4    BP Module : Bio::DB::Taxonomy::entrez
Instances: 7    BP Module : Bio::DB::Taxonomy::flatfile
Instances: 1    BP Module : Bio::Expression::Platform
Instances: 1    BP Module : Bio::SeqIO::genbank
Instances: 22   BP Module : Bio::Taxonomy
Instances: 8    BP Module : Bio::Taxonomy::FactoryI
Instances: 17   BP Module : Bio::Taxonomy::Node
Instances: 20   BP Module : Bio::Taxonomy::Taxon
Instances: 39   BP Module : Bio::Taxonomy::Tree

Hmm, not much.  Almost all hits are within Bio::DB::taxonomy or
Bio::Taxonomy.  The SeqIO::genbank was my change BTW; just haven't tossed
the code yet.  

Therefore, the only one left that would be affected (outside of
Bio::Taxonomy and Bio::DB::Taxonomy) is Allen Day's
Bio::Expression::Platform class, which uses Bio::DB::Taxonomy::entrez to
grab Nodes; that would just be changed over to whatever class you plan on
using.  And that class hasn't been documented at all outside the methods.

Furthermore, judging by the mail list archives the Bio::Taxonomy modules had
very little usage outside of Node.  Jason mentioned on an old post that he
could never get Bio::Taxonomy::Taxon/Tree to work and that Dan Kortschak had
moved  (Dan's last post was in 2003).  Hence the test file comments.   

And you make a good point with Bio::Taxonomy::FactoryI.  

I agree, if the modules haven't served a useful purpose they should be
phased out.

Chris