def test_delete_tree_XML(self):
XML = etree.tostring(etree.parse('data/input/create_matrix.phyml',
parser),
pretty_print=True)
name = "Hill_Davis_2011_1"
trees = obtain_trees(XML)
old_len = len(trees)
new_xml = _swap_tree_in_XML(XML, None, name)
trees = obtain_trees(new_xml)
# loop through all trees, checking them
self.assert_(
trees['Davis_2011_1'] ==
"((A:1.00000,B:1.00000)0.00000:0.00000,(C:1.00000,D:1.00000)0.00000:0.00000)0.00000:0.00000;"
)
self.assert_(len(trees) == old_len - 1)
def test_permute_trees_3(self):
XML = etree.tostring(etree.parse('data/input/permute_trees.phyml',
parser),
pretty_print=True)
trees = obtain_trees(XML)
# contains quoted taxa too
output = permute_tree(trees['Hill_Davis_2011_2'], treefile="newick")
self.assert_(_trees_equal(output, "(A, (B, (C, D, E_E, F, G)));"))
def test_get_all_trees(self):
XML = etree.tostring(etree.parse(single_source_input, parser),
pretty_print=True)
tree = obtain_trees(XML)
# Tree key is source_name_tree_no, so we should have
# Hill_2011_1
expected_tree = '((A_1:1.00000,B_1:1.00000)0.00000:0.00000,F_1:1.00000,E_1:1.00000,(G_1:1.00000,H_1:1.00000)0.00000:0.00000)0.00000:0.00000;'
self.assert_(tree['Hill_2011_1'] == expected_tree)
def test_delete_tree_XML_and_remove_source(self):
XML = etree.tostring(etree.parse('data/input/clean_data.phyml',
parser),
pretty_print=True)
names = ["Hill_2012_1", "Hill_2012_2"]
names.sort(reverse=True)
trees = obtain_trees(XML)
old_len = len(trees)
new_xml = XML
for name in names:
new_xml = _swap_tree_in_XML(new_xml, None, name, delete=True)
trees = obtain_trees(new_xml)
self.assert_(len(trees) == old_len - 2)
# check only one source remains
names = get_all_source_names(new_xml)
self.assert_(len(names) == 1)
self.assert_(names[0] == "Hill_2011")
def test_permute_trees_2(self):
XML = etree.tostring(etree.parse('data/input/permute_trees.phyml',
parser),
pretty_print=True)
trees = obtain_trees(XML)
output = permute_tree(trees['Davis_2011_1'], treefile="newick")
temp_file_handle, temp_file = tempfile.mkstemp(suffix=".new")
f = open(temp_file, "w")
f.write(output)
f.close()
output_trees = import_trees(temp_file)
expected_trees = import_trees("data/output/permute_trees_2.nex")
os.remove(temp_file)
self.assert_(len(output_trees) == len(expected_trees))
for i in range(0, len(output_trees)):
self.assert_(_trees_equal(output_trees[i], expected_trees[i]))
def test_insert_tree_XML(self):
XML = etree.tostring(etree.parse('data/input/create_matrix.phyml',
parser),
pretty_print=True)
name = "Hill_Davis_2011_1"
tree = "(a,b,c);"
new_xml = _swap_tree_in_XML(XML, tree, name)
trees = obtain_trees(new_xml)
# loop through all trees, checking them
self.assert_(
trees['Hill_2011_1'] ==
"((A:1.00000,B:1.00000)0.00000:0.00000,(F:1.00000,E:1.00000)0.00000:0.00000)0.00000:0.00000;"
)
self.assert_(
trees['Davis_2011_1'] ==
"((A:1.00000,B:1.00000)0.00000:0.00000,(C:1.00000,D:1.00000)0.00000:0.00000)0.00000:0.00000;"
)
self.assert_(trees[name] == "(a,b,c);")
def test_clean_data(self):
XML = etree.tostring(etree.parse('data/input/clean_data.phyml',
parser),
pretty_print=True)
XML = clean_data(XML)
trees = obtain_trees(XML)
self.assert_(len(trees) == 2)
expected_trees = {
'Hill_2011_2': '(A,B,(C,D,E));',
'Hill_2011_1': '(A, B, C, (D, E, F));'
}
for t in trees:
self.assert_(_trees_equal(trees[t], expected_trees[t]))
# check only one source remains
names = get_all_source_names(XML)
self.assert_(len(names) == 1)
self.assert_(names[0] == "Hill_2011")
def test_amalgamate_trees_tnt(self):
XML = etree.tostring(etree.parse('data/input/old_stk_input.phyml',
parser),
pretty_print=True)
output_string = amalgamate_trees(XML, format="tnt")
trees = obtain_trees(XML)
# save the file and read it back in. Then we check correct format (i.e. readable) and
# we can check the trees are correct
temp_file_handle, temp_file = tempfile.mkstemp(suffix=".tre")
f = open(temp_file, "w")
f.write(output_string)
f.close()
try:
trees_read = import_trees(temp_file)
except:
self.assert_(False)
# we should get no error
os.remove(temp_file)
self.assert_(len(trees) == len(trees_read))
names = trees.keys()
for i in range(0, len(trees)):
self.assert_(_trees_equal(trees_read[i], trees[names[i]]))
def main():
# do stuff
parser = argparse.ArgumentParser(
prog="supertree_support",
description="Add supertree support measures to your supertree.",
)
parser.add_argument('-v',
'--verbose',
action='store_true',
help="Verbose output: mainly progress reports.",
default=False)
parser.add_argument(
'--measure',
choices=['v', 'v+', 'v-', 'wv', 'wv+', 'wv-'],
default='v+',
help=
"Choose measure to add to the supertree output. The w version weight against the length of the source tree."
)
parser.add_argument('input_file',
metavar='input_file',
nargs=1,
help="Your Phyml")
parser.add_argument('input_supertree',
metavar='input_supertree',
nargs=1,
help="Your supertree")
parser.add_argument(
'output_file',
metavar='output_file',
nargs=1,
help="The output stub for all the various output files")
args = parser.parse_args()
verbose = args.verbose
measure = args.measure
input_file = args.input_file[0]
input_supertree = args.input_supertree[0]
output_stub = args.output_file[0]
if measure == 'wv+':
measure = 10
elif measure == 'wv':
measure = 9
elif measure == 'wv-':
measure = 11
elif measure == 'v+':
measure = 7
elif measure == 'v':
measure = 6
elif measure == 'v-':
measure = 8
else:
print "Unknown metric"
sys.exit(-1)
# see p4.SuperTreeSupport for indices.
# if non stadard decoration, what metric to use
# [S, P, R, Q, WS, WP, V, V+, V-, wV, wV+,wV-]
# 0 1 2 3 4 5 6 7 8 9 10 11
# grab taxa in dataset
if (verbose):
print "Parsing PHYML"
XML = stk.load_phyml(input_file)
taxa = stk.get_all_taxa(XML)
# load supertree
supertree_data = stk.import_tree(input_supertree)
supertree = stk._parse_tree(supertree_data)
terminals = supertree.getAllLeafNames(supertree.root)
if (not len(taxa) == len(terminals)):
# this happens if the supertree has been pruned to remove dodgy taxa
if (verbose):
print "Warning: supertree contains different number of taxa to your input data. Pruning input data"
taxa.sort()
terminals.sort()
delete_me = []
# create subs file
for t in taxa:
if not t in terminals:
delete_me.append(t)
# strip from phyml
if (verbose):
print "Deleting: " + str(len(delete_me)) + " from original " + str(
len(taxa))
try:
XML = stk.substitute_taxa(XML, delete_me, ignoreWarnings=True)
# do we need a clean data to check for non-informative trees here?
except TreeParseError as detail:
msg = "***Error: failed to parse a tree in your data set.\n" + detail.msg
print msg
return
# get all trees from phyml
input_trees = stk.obtain_trees(XML)
source_trees = []
for t in input_trees:
source_trees.append(stk._parse_tree(input_trees[t]))
sts = SuperTreeSupport(input_supertree, source_trees)
sts.doSaveDecoratedTree = True
sts.doStandardDecoration = False
sts.decorationMetric = measure
sts.decoratedFilename = output_stub + '_dec_st.nex'
sts.doSaveIndexTree = True
sts.indexFilename = output_stub + '_index.nex'
sts.csvFilename = output_stub + '_index.csv'
sts.doDrawTree = False
sts.verbose = 1
sts.superTreeSupport()
