def test_matrix2tree(self):
from lingpy.algorithm.clustering import matrix2tree
newick = text_type(self.tmp_path('t'))
matrix2tree([[1, 0], [0, 1]], ['a', 'b'], filename=newick)
assert os.path.exists(newick + '.nwk')
matrix2tree([[1, 0], [0, 1]], ['a', 'b'], tree_calc='upgma')
def test_matrix2tree(self):
from lingpy.algorithm.clustering import matrix2tree
newick = text_type(self.tmp_path('t'))
matrix2tree([[1, 0], [0, 1]], ['a', 'b'], filename=newick)
assert os.path.exists(newick + '.nwk')
matrix2tree([[1, 0], [0, 1]], ['a', 'b'], tree_calc='upgma')
def test_matrix2tree(tmppath, matrix, taxa):
newick = tmppath / 't'
matrix2tree(matrix, taxa, filename=str(newick))
assert newick.parent.joinpath(newick.name + '.nwk').exists()
matrix2tree(matrix, taxa, tree_calc='upgma')
matrix2tree(matrix, taxa, tree_calc='neighbor')
with pytest.raises(ValueError):
matrix2tree(*[matrix, taxa], **{"tree_calc": "dummy"})
def test_matrix2tree(self):
newick = text_type(self.tmp_path('t'))
matrix2tree(self.matrix, self.taxa, filename=newick)
assert os.path.exists(newick + '.nwk')
matrix2tree(self.matrix, self.taxa, tree_calc='upgma')
matrix2tree(self.matrix, self.taxa, tree_calc='neighbor')
assert_raises(ValueError, matrix2tree, *[self.matrix, self.taxa],
**{"tree_calc": "dummy"})
def test_matrix2tree(self):
from lingpy.algorithm.clustering import matrix2tree
newick = text_type(self.tmp_path('t'))
matrix2tree(self.matrix, self.taxa, filename=newick)
assert os.path.exists(newick + '.nwk')
matrix2tree(self.matrix, self.taxa, tree_calc='upgma')
matrix2tree(self.matrix, self.taxa, tree_calc='neighbor')
assert_raises(ValueError, matrix2tree, *[self.matrix, self.taxa],
**{"tree_calc": "dummy"})
def calculate_data(wordlist,
data,
taxa='taxa',
concepts='concepts',
ref='cogid',
**keywords):
"""
Manipulate a wordlist object by adding different kinds of data.
Parameters
----------
data : str
The type of data that shall be calculated. Currently supports
* "tree": calculate a reference tree based on shared cognates
* "dst": get distances between taxa based on shared cognates
* "cluster": cluster the taxa into groups using different methods
"""
logger = log.get_logger()
util.setdefaults(keywords,
distances=False,
tree_calc="upgma",
cluster="upgma",
force=False,
threshold=0.5,
cluster_method='upgma')
# get taxa for current calculation
these_taxa = eval('wordlist.' + taxa)
# calculate distances
if data in ['distances', 'dst']:
wordlist._meta['distances'] = wl2dst(wordlist, taxa, concepts, ref,
**keywords)
elif data in ['diversity', 'div']:
etd = wordlist.get_etymdict(ref=ref)
wordlist._meta['diversity'] = \
(len(etd) - wordlist.height) / (len(wordlist) - wordlist.height)
elif data in ['tre', 'tree', 'nwk']:
if 'distances' not in wordlist._meta:
wordlist._meta['distances'] = \
wl2dst(wordlist, taxa, concepts, ref, **keywords)
distances = wordlist._meta['distances']
if 'tree' in wordlist._meta and not keywords['force']:
logger.warning("Reference tree has already been calculated, "
"force overwrite by "
"setting 'force' to 'True'.")
return
wordlist._meta['tree'] = clustering.matrix2tree(
distances, these_taxa, keywords['tree_calc'],
keywords['distances'])
elif data in ['groups', 'cluster']:
if 'distances' not in wordlist._meta:
distances = wl2dst(wordlist, taxa, concepts, ref, **keywords)
else:
distances = wordlist._meta['distances']
if 'groups' in wordlist._meta and not keywords['force']:
logger.warning("Distance matrix has already been calculated, "
"force overwrite by "
"setting 'force' to 'True'.")
return
wordlist._meta['groups'] = clustering.matrix2groups(
keywords['threshold'], distances, these_taxa,
keywords['cluster_method'])
log.info("Successfully calculated {0}.".format(data))
def calculate_data(
wordlist,
data,
taxa='taxa',
concepts='concepts',
ref='cogid',
**keywords):
"""
Manipulate a wordlist object by adding different kinds of data.
Parameters
----------
data : str
The type of data that shall be calculated. Currently supports
* "tree": calculate a reference tree based on shared cognates
* "dst": get distances between taxa based on shared cognates
* "cluster": cluster the taxa into groups using different methods
"""
logger = log.get_logger()
util.setdefaults(
keywords,
distances=False,
tree_calc="upgma",
cluster="upgma",
force=False,
threshold=0.5,
cluster_method='upgma')
# get taxa for current calculation
these_taxa = eval('wordlist.' + taxa)
# calculate distances
if data in ['distances', 'dst']:
wordlist._meta['distances'] = wl2dst(
wordlist, taxa, concepts, ref, **keywords)
elif data in ['diversity', 'div']:
etd = wordlist.get_etymdict(ref=ref)
wordlist._meta['diversity'] = \
(len(etd) - wordlist.height) / (len(wordlist) - wordlist.height)
elif data in ['tre', 'tree', 'nwk']:
if 'distances' not in wordlist._meta:
wordlist._meta['distances'] = \
wl2dst(wordlist, taxa, concepts, ref, **keywords)
distances = wordlist._meta['distances']
if 'tree' in wordlist._meta and not keywords['force']:
logger.warn(
"Reference tree has already been calculated, "
"force overwrite by "
"setting 'force' to 'True'.")
return
wordlist._meta['tree'] = clustering.matrix2tree(
distances, these_taxa, keywords['tree_calc'],
keywords['distances'])
elif data in ['groups', 'cluster']:
if 'distances' not in wordlist._meta:
distances = wl2dst(wordlist, taxa, concepts, ref, **keywords)
else:
distances = wordlist._meta['distances']
if 'groups' in wordlist._meta and not keywords['force']:
logger.warn(
"Distance matrix has already been calculated, "
"force overwrite by "
"setting 'force' to 'True'.")
return
wordlist._meta['groups'] = clustering.matrix2groups(
keywords['threshold'], distances, these_taxa,
keywords['cluster_method'])
log.info("Successfully calculated {0}.".format(data))