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

[Sendu Bala]

Chris Fields wrote:
> How 'broken', exactly, is Bio::Taxonomy?

Its certainly usable as-is, but there are some gotchas.

# It has an acknowledged weakness of not coping with multiple ranks of 
the same name (notably 'no rank').

# You can't have 2 nodes with the same rank (so can only build a single 
lineage, not a whole menagerie).

# You must supply a list of all your rank names correctly ordered before 
you can add any nodes (or trust that the default list is satisfactory - 
it won't be if you have just a single 'no rank').

# You simply don't need it if you have Bio::Taxonomy::Nodes with 
db_handle set, or Bio::Taxonomy::Taxons.

In my opinion, the burden is just too great for this ever to have been a 
'fun' module to use. It was only required so that people could manually 
create their own Bio::Taxonomy::Nodes and form a lineage without a database.


> Don't know if it has been attempted but it would be nice to have a way for
> building taxonomic trees from Node/Taxon information using a Taxonomy-like
> container object.  I like the way NCBI does something along these lines with
> BLAST output now.

Not really sure what you mean. I don't think you'd require a container 
object to do any particular task. Can you clarify?