[Phylobase-devl] IDs vs. labels (support for nexml format)

[Aaron Mackey]

Last week I spent some time working on "read.nexml" and "write.nexml"
functions to support IO in nexml format (see: www.nexml.org), a modern
replacement for NEXUS.

These functions are now working (minimally), but they have a disturbing
feature: while nexml trees have both an identifier and (optional) label for
each node in a tree, phylo4 objects have only one "slot".  Thus, a tree read
from nexml cannot be fully represented by a phylo4 object (and so cannot be
accurately "round-tripped".

Is there any ability for additional "annotation" on phylo4 nodes?  I realize
I can co-opt a phylo4d data.frame for this purpose, but that seems to mix
the character data with the "metainformation" of a node annotation.  Could
the phylo4 object model be simply extended to have an identifier and
(optional) label slot?

Thanks for your thoughts,

-Aaron
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.r-forge.r-project.org/pipermail/phylobase-devl/attachments/20080528/d3f6351b/attachment.htm