DendroPy 3.10.0

[Jeet Sukumaran]

Version 3.10.0 of DendroPy has been released:

http://pypi.python.org/pypi/DendroPy

One major bug fix [under some circumstances the log-probability of the
coalescent of a gene tree was calculated incorrectly], and a few new
features related to NEXUS-format input / output [making life easier
when combining data from multiple files or split a character matrix in
separate files].

EVERYONE IS HIGHLY ENCOURAGED TO UPGRADE TO THIS VERSION DUE TO THE
POSSIBILITY OF AN INCORRECT COALESCENT PROBABILITY IN THE PREVIOUS
VERSION!

If you used 'pip' to install, then::

$ sudo pip install --upgrade dendropy

If you used 'setuptools', then::

$ sudo easy_install -U dendropy

Otherwise simply download the archive and run::

$ sudo setup.py install

Of course, the true geeks and hackers amongst you will have cloned the
GitHub repo, and are running 'live' out of that, so you would just
need to::

$ git pull

Documentation for the new features are lagging, as usual, but here are
some quick hints.

Quickly get a collection splits present on one tree but not on another
by::

tree.find_missing_splits(other)

Create a merged data set by::

d = dendropy.DnaCharacterMatrix.concatenate_from_paths(["d1.nex",
"d2.nex". "d3.nex"],
"nexus")

This will create a single character matrix, 'd', consisting of all the
matrices given in the list concatenated. Each source matrix is assumed
to be an alignment, i.e., all the same length. In addtion, character
subset objects will automatically be created, with one subset per
source data file, to track the positions (columns) of the matrix and
their original source.

This means that you can export the original source file via using the
'CharacterMatrix.export_character_subset()' method. Note that to use
the latter you will need to get a reference to relevant character
subset.

Detailed change log:

http://packages.python.org/DendroPy/changes.html