def setUp(self):
self.taxon_set = dendropy.TaxonSet([
'a1', 'a2', 'a3', 'a4', 'b1', 'b2', 'b3', 'b4', 'c1', 'c2', 'c2',
'c3', 'd1', 'a5', 'a6', 'd2', 'd3'
])
self.membership_func = lambda x: x.label[0]
self.membership_dict = {}
for t in self.taxon_set:
self.membership_dict[t] = t.label[0]
t.subset_id = t.label[0]
self.membership_lists = [[
self.taxon_set[0], self.taxon_set[1], self.taxon_set[2],
self.taxon_set[3], self.taxon_set[13], self.taxon_set[14]
],
[
self.taxon_set[4], self.taxon_set[5],
self.taxon_set[6], self.taxon_set[7]
],
[
self.taxon_set[8], self.taxon_set[9],
self.taxon_set[10], self.taxon_set[11]
],
[
self.taxon_set[12], self.taxon_set[15],
self.taxon_set[16]
]]
self.label_map = ['a', 'b', 'c', 'd']
self.expected_sets = set([dendropy.TaxonSet(s, label=self.label_map[i]) \
for i, s in enumerate(self.membership_lists)])
self.expected_dict = {}
for s in self.expected_sets:
self.expected_dict[self.membership_dict[s[0]]] = s
def setUp(self):
self.domain_taxa = dendropy.TaxonSet([
'a1', 'a2', 'a3', 'a4', 'b1', 'b2', 'b3', 'b4', 'c1', 'c2', 'c2',
'c3', 'd1', 'a5', 'a6', 'd2', 'd3'
])
self.range_taxa = dendropy.TaxonSet([
'A',
'B',
'C',
'D',
])
self.domain_taxa.lock()
self.range_taxa.lock()
self.mapping_func = lambda x: self.range_taxa.require_taxon(
label=x.label[0].upper())
self.mapping_dict = {}
for t in self.domain_taxa:
self.mapping_dict[t] = self.mapping_func(t)
t.containing_taxa = self.mapping_dict[t]
self.expected_forward_label_map = {
'a1': 'A',
'a2': 'A',
'a3': 'A',
'a4': 'A',
'a5': 'A',
'a6': 'A',
'b1': 'B',
'b2': 'B',
'b3': 'B',
'b4': 'B',
'c1': 'C',
'c2': 'C',
'c3': 'C',
'c4': 'C',
'd1': 'D',
'd2': 'D',
'd3': 'D',
}
self.expected_backward_label_map = {
'A': set(['a1', 'a2', 'a3', 'a4', 'a5', 'a6']),
'B': set([
'b1',
'b2',
'b3',
'b4',
]),
'C': set([
'c1',
'c2',
'c2',
'c3',
]),
'D': set(['d1', 'd2', 'd3'])
}
def setUp(self):
self.tb1 = dendropy.TaxonSet(label="TI1")
for i in range(1, 11):
self.tb1.new_taxon(label="T%02d" % i)
self.cb1 = dendropy.DnaCharacterMatrix(taxon_set=self.tb1,
label="TI1, CA1")
for t in self.tb1:
self.cb1.append_taxon_sequence(t, state_symbols="AAAAAAAAAA")
self.tb2 = dendropy.TaxonSet(label="TI2")
for i in range(1, 21):
self.tb2.new_taxon(label="T%02d" % i)
self.cb2 = dendropy.DnaCharacterMatrix(taxon_set=self.tb2,
label="TI2, CA2")
for t in self.tb2:
self.cb2.append_taxon_sequence(t, state_symbols="CCCCCCCCCC")
def phylo_p(self):
message(self,
'Generating random genotypes (%d x %d)' % (self.n, self.l))
# load dendropy
import dendropy
# make alignment from random tree
tree = dendropy.treesim.pure_kingman(
dendropy.TaxonSet(map(str, range(self.n))))
seqs = [
si for si in dendropy.seqsim.generate_hky_dataset(
seq_len=self.l * 10, tree_model=tree,
mutation_rate=self.u).as_string('fasta').split('\n')
if si != '' and si[0] != '>'
]
# count k-morphic sites
counts = np.array([
len(set([seqs[i][j] for i in range(self.n)]))
for j in range(self.l * 10)
])
# look at 2-3 morphic sites
sites = [i for i in range(len(counts)) if 1 < counts[i] < 4][:self.l]
# construct p
self.p = np.array([[bps[seqs[i][j]] for j in sites]
for i in range(self.n)])
return self
def rf_weighted(tree_object1, tree_object2):
tree_newick1 = tree_object1.newick(tree_object1.root) + ";"
tree_newick2 = tree_object2.newick(tree_object2.root) + ";"
#print(tree_newick1)
#print(tree_newick2)
version = dendropy.__version__.split(".")[0]
if version == '4':
taxa = dendropy.TaxonNamespace() #set taxa same for all
tree1 = dendropy.Tree.get(data=tree_newick1,
schema='newick',
taxon_namespace=taxa,
rooting='force-rooted')
tree2 = dendropy.Tree.get(data=tree_newick2,
schema='newick',
taxon_namespace=taxa,
rooting='force-rooted')
elif version == '3':
taxa = dendropy.TaxonSet() #set taxa same for all
tree1 = dendropy.Tree.get(data=tree_newick1,
schema='newick',
taxon_set=taxa,
rooting='force-rooted')
tree2 = dendropy.Tree.get(data=tree_newick2,
schema='newick',
taxon_set=taxa,
rooting='force-rooted')
tree1.encode_bipartitions()
tree2.encode_bipartitions()
dist = dendropy.calculate.treecompare.weighted_robinson_foulds_distance(
tree1, tree2)
return dist
def root_trees(): global treefname taxa = dp.TaxonSet() treelist = dp.TreeList() treelist.read_from_path(treefname, schema="newick", taxon_set=taxa) global outgroup_taxon_names outgroup_taxa = list() for name in outgroup_taxon_names: for t in taxa: print t.label if t.label == name: outgroup_taxa.append(t) print outgroup_taxa for tree in treelist: rootnode = tree.mrca(taxa=outgroup_taxa) tree.reroot_at_edge(rootnode.edge, length1 = rootnode.edge_length / 2 , length2 = rootnode.edge_length / 2, update_splits = True) tree.print_plot() outfile = open(treefname + ".rooted", "wb") treelist.write(outfile,schema="newick", edge_lengths = True) rooted_trees_fname = outfile.name outfile.close()
def getTree(self, size, birthParam):
tree = treesim.birth_death(birth_rate=birthParam, death_rate=0, taxon_set=dendropy.TaxonSet(self.fullTaxonSet[0:size]))
#tree.deroot()
#print(tree)
#randomize slightly
#self.rescaleTree(tree,1.0)
return tree
def setUp(self):
self.labels = []
for idx in xrange(10):
self.labels.append("T%d" % (idx + 1))
self.taxon_set = dendropy.TaxonSet()
for label in self.labels:
self.taxon_set.new_taxon(label=label)
def testBindToSpecifiedTaxonSet(self):
d = dendropy.DataSet()
t = dendropy.TaxonSet()
d.attach_taxon_set(t)
self.assertEqual(len(d.taxon_sets), 1)
self.assertIs(d.taxon_sets[0], d.attached_taxon_set)
self.assertIs(d.attached_taxon_set, t)
d.read_from_path(
pathmap.mixed_source_path('reference_single_taxonset_dataset.nex'),
"nexus")
self.assertEqual(len(d.taxon_sets), 1)
self.assertEqual(len(d.taxon_sets[0]), 33)
d.read_from_path(pathmap.tree_source_path('pythonidae.mle.nex'),
"nexus")
self.assertEqual(len(d.taxon_sets), 1)
self.assertEqual(len(d.taxon_sets[0]), 33)
d.read_from_path(
pathmap.tree_source_path('pythonidae.reference-trees.newick'),
"newick")
self.assertEqual(len(d.taxon_sets), 1)
self.assertEqual(len(d.taxon_sets[0]), 33)
d.detach_taxon_set()
d.read_from_path(
pathmap.char_source_path('caenophidia_mos.chars.fasta'),
"proteinfasta")
self.assertEqual(len(d.taxon_sets), 2)
self.assertEqual(len(d.taxon_sets[0]), 33)
self.assertEqual(len(d.taxon_sets[1]), 114)
def testSimple(self):
taxa = dendropy.TaxonSet()
tree1_str = "[&U] (A,(B,(C,(D,E))));"
tree2_str = "[&U] (B,(C,(D,(A,E))));"
tree3_str = "[&U] (D,(A,(B,(C,E))));"
tree4_str = "[&U] (C,(D,(A,(B,E))));"
tree5_str = "[&U] (A,(E,(B,(C,D))));"
all_tree_strs = [tree1_str, tree2_str, tree3_str, tree4_str, tree5_str]
weights = [8, 5, 4, 2, 1]
test_tree_strs = []
for idx, tree_str in enumerate(all_tree_strs):
test_tree_strs.extend([tree_str] * weights[idx])
test_trees = dendropy.TreeList.get_from_string(
"\n".join(test_tree_strs), 'newick', taxon_set=taxa)
tc = treesum.TopologyCounter()
expected_freq_values = [float(i) / sum(weights) for i in weights]
expected_trees = dendropy.TreeList.get_from_string(
"\n".join(all_tree_strs), 'newick', taxon_set=taxa)
for tree in test_trees:
tc.count(tree)
result_tree_freqs = tc.calc_tree_freqs(taxon_set=taxa)
for idx, (result_tree,
result_freq) in enumerate(result_tree_freqs.items()):
expected_tree = expected_trees[idx]
expected_tree.update_splits()
expected_freq = expected_freq_values[idx]
expected_count = weights[idx]
self.assertEqual(
result_tree.symmetric_difference(expected_tree), 0,
"%s != %s" % (result_tree.as_string('newick'),
expected_tree.as_string('newick')))
self.assertAlmostEqual(result_freq[0], expected_count)
self.assertAlmostEqual(result_freq[1], expected_freq)
def convert_to_dendropy_trees(trees):
taxa = dpy.TaxonSet()
dpy_tree_list = [
dpy.Tree.get_from_string(tree.newick, 'newick', taxon_set=taxa)
for tree in trees
]
return dpy_tree_list
def setUp(self):
tree_str = "[&R] ((((H**o:0.21,Bogus1:0.23,Pongo:0.21)N1:0.28,Bogus2:0.49,Macaca:0.49)N2:0.13,Bogus3:0.62,Ateles:0.62)N3:0.38,Galago:1.00)N4:0.0;"
data_str = """
#NEXUS
BEGIN DATA;
DIMENSIONS NTAX=8 NCHAR=2;
FORMAT DATATYPE = CONTINUOUS GAP = - MISSING = ?;
MATRIX
H**o 4.09434 4.74493
Pongo 3.61092 3.33220
Macaca 2.37024 3.36730
Ateles 2.02815 2.89037
Galago -1.46968 2.30259
Bogus1 2.15 2.15
Bogus2 2.15 2.15
Bogus3 2.15 2.15
;
END;
"""
taxa = dendropy.TaxonSet()
self.tree = dendropy.Tree.get_from_string(tree_str,
'newick',
taxon_set=taxa)
self.char_matrix = dendropy.ContinuousCharacterMatrix.get_from_string(
data_str, 'nexus', taxon_set=taxa)
def testTreeFromTreeDistinctTaxa(self):
taxa = dendropy.TaxonSet()
tree2 = dendropy.Tree(self.tree1, taxon_set=taxa)
self.assertIs(tree2.taxon_set, taxa)
self.assertIsNot(tree2.taxon_set, self.tree1.taxon_set)
self.assertDistinctButEqual(self.tree1,
tree2,
distinct_taxa=True,
equal_oids=False)
def compare_trees(expected, estimated):
# assert(estimated.euclidean_distance(expected)<= branch lengths are hard to test. TODO figure out how.
taxon_namespace = dendropy.TaxonSet()
exp_tree = dendropy.Tree.get_from_path(expected,
"newick",
taxon_set=taxon_namespace)
est_tree = dendropy.Tree.get_from_path(estimated,
"nexus",
taxon_set=taxon_namespace)
return (est_tree.symmetric_difference(exp_tree))
def read_tree(self):
"""Reads in a tree from a file, arbitrarily resolves poltomies if present,
strips leading [&U] and writes out to outputdir/simtree.tre"""
self._treeread = 1
if not self._madeout:
self.make_output()
#import tree from path
if dendropy.__version__.startswith('4'):
taxa = dendropy.TaxonNamespace()
try:
tree = dendropy.Tree.get_from_path(self.get_arg('treepath'),
self.treetype,
taxon_namespace=taxa)
except:
sys.stderr.write(
"Problems reading the tree - is it in proper newick or nexus format?\n"
)
else:
taxa = dendropy.TaxonSet()
try:
tree = dendropy.Tree.get_from_path(self.get_arg('treepath'),
self.treetype,
taxon_set=taxa)
except:
sys.stderr.write(
"Problems reading the tree - is it in proper newick or nexus format?\n"
)
if tree.length() == 0:
sys.stderr.write(
"TTR requires branch lengths. Branch lengths appear to be missing (treelength = 0). Exiting.\n"
)
self._exit_handler()
self.seqnames = taxa.labels()
f = open('labels.txt', 'w')
for name in self.seqnames:
f.write(self.prefix + name + '\n')
f.close()
if not self.get_arg('base_name') in self.seqnames:
sys.stderr.write(
"base genome name {} is not in tree. Exiting.\n".format(
self.get_arg('base_name')))
self._exit_handler()
tree.resolve_polytomies()
self.seqgen_scaler = float((1.0 / tree.length()) * 10)
# sys.stdout.write("scaler is {}\n".format(self.seqgen_scaler))
if tree.length() >= 1:
sys.stderr.write(
"WARNING: Tree length is high- scale down tree or expect high multiple hits/homoplasy!\n"
)
self.outtree = "{}/simtree.tre".format(self.outd)
tree.write_to_path(self.outtree,
schema='newick',
suppress_internal_node_labels=True,
suppress_rooting=True)
sys.stdout.write("Tree read\n")
def read_tree(self):
"""Reads in a tree from a file, arbitrarily resolves poltomies if present,
strips leading [&U] and writes out to outputdir/simtree.tre"""
self._treeread = 1
if not self._madeout:
self.make_output()
#import tree from path
if dendropy.__version__.startswith('4'):
taxa = dendropy.TaxonNamespace()
try:
tree = dendropy.Tree.get_from_path(self.get_arg('treepath'), self.treetype, taxon_namespace=taxa, preserve_underscores=True)
except:
sys.stderr.write("Problems reading the tree - is it in proper newick or nexus format?\n")
self._exit_handler()
else:
taxa = dendropy.TaxonSet()
try:
tree = dendropy.Tree.get_from_path(self.get_arg('treepath'), self.treetype, taxon_set=taxa, preserve_underscores=True)
except:
sys.stderr.write("Problems reading the tree - is it in proper newick or nexus format?\n")
self._exit_handler()
if tree.length() == 0:
sys.stderr.write("TTR requires branch lengths. Branch lengths appear to be missing (treelength = 0). Exiting.\n")
self._exit_handler()
self.seqnames = taxa.labels()
self.base_name = self.get_arg('base_name')
if self.base_name not in self.seqnames:
sys.stderr.write("base genome name {} is not in tree. Exiting.\n".format(self.base_name))
self._exit_handler()
tree.resolve_polytomies()
tree_len = tree.length()
expected_tree_len = float(self.nsnp)/self.genlen
for edge in tree.postorder_edge_iter():
if edge.length is None:
edge.length = 0
else:
edge.length = (float(edge.length)/tree_len) * expected_tree_len
assert -0.001 < expected_tree_len - tree.length() < 0.001
self.scaledouttree = "{}/scaledtree.tre".format(self.outd)
tree.write_to_path(self.scaledouttree,
schema='newick',
suppress_internal_node_labels=True,
suppress_rooting=True)
self.scaled_tree_newick = tree.as_string(schema='newick', real_value_format_specifier='.15f')
if expected_tree_len < 0.01: #scale up tree length so generate mutations in seqgen without a million invariant sites.
stretch = 0.01/expected_tree_len
for edge in tree.postorder_edge_iter():
if edge.length is None:
edge.length = 0
else:
edge.length = edge.length * stretch
self.outtree = "{}/simtree.tre".format(self.outd)
tree.write_to_path(self.outtree, schema='newick', suppress_internal_node_labels=True, suppress_rooting=True)
sys.stdout.write("Tree read\n")
def assertDistinctButEqualTree(self, tree1, tree2, **kwargs):
otaxa = tree1.taxon_set
ts = dendropy.TaxonSet()
tree1.reindex_taxa(ts, clear=True)
tree2.reindex_taxa(ts)
self.assertIs(tree1.taxon_set, tree2.taxon_set)
self.assertIsNot(tree1.taxon_set, otaxa)
self.assertDistinctButEqual(tree1.taxon_set, otaxa, **kwargs)
treesplit.encode_splits(tree1)
treesplit.encode_splits(tree2)
rfdist = treecalc.robinson_foulds_distance(tree1, tree2)
self.assertAlmostEqual(rfdist, 0)
def testFittedDeepCoalCounting(self):
taxa = dendropy.TaxonSet()
gene_trees = dendropy.TreeList.get_from_string("""
[&R] (A,(B,(C,D))); [&R] ((A,C),(B,D)); [&R] (C,(A,(B,D)));
""",
"newick",
taxon_set=taxa)
species_trees = dendropy.TreeList.get_from_string("""
[&R] (A,(B,(C,D)));
[&R] (A,(C,(B,D)));
[&R] (A,(D,(C,B)));
[&R] (B,(A,(C,D)));
[&R] (B,(C,(A,D)));
[&R] (B,(D,(C,A)));
[&R] (C,(A,(B,D)));
[&R] (C,(B,(A,D)));
[&R] (C,(D,(B,A)));
[&R] (D,(A,(B,C)));
[&R] (D,(B,(A,C)));
[&R] (D,(C,(B,A)));
[&R] ((A,B),(C,D));
[&R] ((A,C),(B,D));
[&R] ((A,D),(C,B));
""",
"NEWICK",
taxon_set=taxa)
# expected results, for each gene tree / species tree pairing, with
# cycling through species trees for each gene tree
expected_deep_coalescences = [
0, 1, 1, 1, 2, 2, 3, 3, 3, 3, 3, 3, 1, 2, 2, 2, 1, 2, 2, 2, 1, 1,
2, 2, 2, 1, 2, 2, 0, 2, 2, 1, 2, 3, 3, 3, 0, 1, 1, 3, 3, 3, 2, 1, 2
]
assert len(
expected_deep_coalescences) == len(gene_trees) * len(species_trees)
for t in gene_trees + species_trees:
t.update_splits()
idx = 0
_LOG.info("Species\t\tGene\t\tDC\t\tExp.DC\t\tDiff")
for gt in gene_trees:
for st in species_trees:
dc = reconcile.reconciliation_discordance(gt, st)
_LOG.info("%s\t\t%s\t\t%s\t\t%s\t\t%s" %
(st.compose_newick(), gt.compose_newick(), dc,
expected_deep_coalescences[idx],
dc - expected_deep_coalescences[idx]))
assert dc == expected_deep_coalescences[idx]
idx += 1
def setUp(self):
self.taxon_set = dendropy.TaxonSet()
self.support_trees = dendropy.TreeList.get_from_path(
pathmap.tree_source_path("primates.beast.mcmc.trees"),
"nexus",
taxon_set=self.taxon_set,
tree_offset=40)
self.split_distribution = treesplit.SplitDistribution(
taxon_set=self.taxon_set)
self.split_distribution.is_rooted = True
self.split_distribution.ignore_node_ages = False
for tree in self.support_trees:
tree.update_splits()
self.split_distribution.count_splits_on_tree(tree)
def testTaxaQuerying(self):
ts = dendropy.TaxonSet(self.labels)
self.assertTrue(ts.has_taxa(labels=self.labels))
self.assertTrue(ts.has_taxa(taxa=ts))
self.assertFalse(ts.has_taxa(labels=self.labels + ["k"]))
k = ts.new_taxon(label="k")
self.assertTrue(ts.has_taxa(taxa=[k]))
self.assertTrue(ts.has_taxon(label="k"))
self.assertTrue(ts.has_taxa(labels=self.labels + ["k"]))
j = dendropy.Taxon(label="j")
ts.add_taxon(j)
self.assertTrue(ts.has_taxa(taxa=[j]))
self.assertTrue(ts.has_taxon(label="j"))
self.assertTrue(ts.has_taxa(labels=self.labels + ["j"]))
self.assertFalse(ts.has_taxon(taxon=dendropy.Taxon()))
for label in self.labels:
self.assertTrue(ts.has_taxon(label=label))
def testMixedNexusAndNewickDistinctTaxa(self):
filenames = [
datagen.reference_trees_filename(schema="newick"),
datagen.reference_trees_filename(schema="nexus"),
datagen.reference_trees_filename(schema="newick"),
datagen.reference_trees_filename(schema="nexus")
]
filepaths = [pathmap.tree_source_path(f) for f in filenames]
taxon_set = dendropy.TaxonSet()
for idx, test_tree in enumerate(
dataio.multi_tree_source_iter(filepaths,
schema="nexus/newick",
taxon_set=taxon_set)):
self.assertDistinctButEqualTree(self.next_ref_tree(),
test_tree,
distinct_taxa=True,
ignore_taxon_order=True)
self.assertEqual(idx, 43)
def runTest(self):
taxon_set = dendropy.TaxonSet([str(i+1) for i in range(5)])
tree_list = dendropy.TreeList(
stream=StringIO("""
(5,((4,3),2),1);
(5,(4,3,2),1);
(5,((4,3),2),1);
(5,(4,3),2,1);
(5,((4,3),2),1);
(5,4,3,2,1);
"""),
schema="newick",
taxon_set=taxon_set)
tree = tree_list[0]
expected_tree = tree_list[1]
treesplit.encode_splits(tree)
all_cm = tree.seed_node.edge.split_bitmask
split_to_target = 0xA
treemanip.collapse_conflicting(tree.seed_node, split_to_target, all_cm)
treesplit.encode_splits(tree)
treesplit.encode_splits(expected_tree)
self.assertEqual(treecalc.symmetric_difference(tree, expected_tree), 0)
tree = tree_list[2]
expected_tree = tree_list[3]
treesplit.encode_splits(tree)
all_cm = tree.seed_node.edge.split_bitmask
split_to_target = 0x3
treemanip.collapse_conflicting(tree.seed_node, split_to_target, all_cm)
treesplit.encode_splits(tree)
treesplit.encode_splits(expected_tree)
self.assertEqual(treecalc.symmetric_difference(tree, expected_tree), 0)
tree = tree_list[4]
expected_tree = tree_list[5]
treesplit.encode_splits(tree)
all_cm = tree.seed_node.edge.split_bitmask
split_to_target = 0x5
treemanip.collapse_conflicting(tree.seed_node, split_to_target, all_cm)
treesplit.encode_splits(tree)
treesplit.encode_splits(expected_tree)
self.assertEqual(treecalc.symmetric_difference(tree, expected_tree), 0)
def generate(self, trees, dataset=None, taxon_set=None, **kwargs):
args = self._compose_arguments()
tree_inputf = self.get_tempfile()
trees.write_to_path(tree_inputf.name,
"newick",
suppress_rooting=True,
suppress_internal_node_labels=True)
tree_inputf.flush()
args.append(tree_inputf.name)
run = subprocess.Popen(args,
stdout=subprocess.PIPE,
stderr=subprocess.PIPE)
stdout, stderr = run.communicate()
if taxon_set is None:
taxon_set = dendropy.TaxonSet()
if dataset is None:
dataset = dendropy.DataSet(taxon_set=taxon_set, **kwargs)
results = StringIO(stdout)
dataset.read(results, "nexus")
return dataset
def testSummarizeNodeAgesOnMCCT(self):
"""
SumTrees: summarizing node ages on MCCT topology.
"""
if runlevel.is_test_enabled(runlevel.EXHAUSTIVE, _LOG,
self.__class__.__name__):
path_to_src = pathmap.tree_source_path("primates.beast.mcmc.trees")
path_to_target = pathmap.tree_source_path(
"primates.beast.mcct.noedgelens.tree")
args = [
"-b", "40", "-e", "mean-age", "-t", path_to_target, path_to_src
]
retcode, stdout, stderr = self.execute_sumtrees(args)
self.assertEqual(retcode, 0)
taxa = dendropy.TaxonSet()
exp_tree = dendropy.Tree.get_from_path(
pathmap.tree_source_path("primates.beast.mcct.meanh.tre"),
"nexus",
taxon_set=taxa)
obs_tree = dendropy.Tree.get_from_string(stdout,
"nexus",
taxon_set=taxa)
exp_tree.update_splits()
exp_tree.calc_node_ages()
obs_tree.update_splits()
obs_tree.calc_node_ages()
self.assertEqual(exp_tree.split_edges.keys(),
obs_tree.split_edges.keys())
splits = exp_tree.split_edges.keys()
for split in splits:
exp_edge = exp_tree.split_edges[split]
obs_edge = obs_tree.split_edges[split]
self.assertAlmostEqual(obs_edge.head_node.age,
exp_edge.head_node.age)
else:
_LOG.info(
"Skipping test (set 'DENDROPY_TESTING_LEVEL=EXHAUSTIVE' to run)"
)
def parse_taxon_set(self):
"""
Given PAUP* output that includes a taxon listing as produced by
`stage_list_taxa`, this parses out and returns a taxon block.
"""
taxlabels = []
taxinfo_pattern = re.compile('\s*(\d+) (.*)\s+\-')
idx = 0
for line in self.output:
idx += 1
if line == "TAXON LIST BEGIN":
break
for line in self.output[idx:]:
if line == "TAXON LIST END":
break
ti_match = taxinfo_pattern.match(line)
if ti_match:
taxlabels.append(ti_match.group(2).strip())
taxon_set = dendropy.TaxonSet()
for taxlabel in taxlabels:
taxon_set.new_taxon(label=taxlabel)
return taxon_set
def setUp(self):
tree_str = "[&R] ((((H**o:0.21,Pongo:0.21)N1:0.28,Macaca:0.49)N2:0.13,Ateles:0.62)N3:0.38,Galago:1.00)N4:0.0;"
data_str = """
#NEXUS
BEGIN DATA;
DIMENSIONS NTAX=5 NCHAR=2;
FORMAT DATATYPE = CONTINUOUS GAP = - MISSING = ?;
MATRIX
H**o 4.09434 4.74493
Pongo 3.61092 3.33220
Macaca 2.37024 3.36730
Ateles 2.02815 2.89037
Galago -1.46968 2.30259
;
END;
"""
taxa = dendropy.TaxonSet()
self.tree = dendropy.Tree.get_from_string(tree_str,
'newick',
taxon_set=taxa)
self.char_matrix = dendropy.ContinuousCharacterMatrix.get_from_string(
data_str, 'nexus', taxon_set=taxa)
self.pic = continuous.PhylogeneticIndependentConstrasts(
tree=self.tree, char_matrix=self.char_matrix)
self.expected_vals = []
self.expected_vals.append({
# state, corrected edge length, contrast, contrast_var
"N1": (3.852630000, 0.385000000, 0.483420000, 0.420000000),
"N2": (3.200378400, 0.345600000, 1.482390000, 0.875000000),
"N3": (2.780823579, 0.601905551, 1.172228400, 0.965600000),
"N4": (1.183724613, 0.375743470, 4.250503579, 1.601905551),
})
self.expected_vals.append({
# state, corrected edge length, contrast, contrast_var
"N1": (4.038565000, 0.385000000, 1.412730000, 0.420000000),
"N2": (3.743208400, 0.345600000, 0.671265000, 0.875000000),
"N3": (3.437967150, 0.601905551, 0.852838400, 0.965600000),
"N4": (3.011356599, 0.375743470, 1.135377150, 1.601905551),
})
def rf_unweighted(tree_object1, tree_object2, normalized='F'):
'''
Gives RF=rf_unweighted[0] and normalized RF=rf_unweighted[1]
'''
tree_newick1 = tree_object1.newick(tree_object1.root) + ";"
tree_newick2 = tree_object2.newick(tree_object2.root) + ";"
#print(tree_newick1)
#print(tree_newick2)
version = dendropy.__version__.split(".")[0]
if version == '4':
taxa = dendropy.TaxonNamespace() #set taxa same for all
tree1 = dendropy.Tree.get(data=tree_newick1,
schema='newick',
taxon_namespace=taxa,
rooting='force-rooted')
tree2 = dendropy.Tree.get(data=tree_newick2,
schema='newick',
taxon_namespace=taxa,
rooting='force-rooted')
elif version == '3':
taxa = dendropy.TaxonSet() #set taxa same for all
tree1 = dendropy.Tree.get(data=tree_newick1,
schema='newick',
taxon_set=taxa,
rooting='force-rooted')
tree2 = dendropy.Tree.get(data=tree_newick2,
schema='newick',
taxon_set=taxa,
rooting='force-rooted')
tree1.encode_bipartitions()
tree2.encode_bipartitions()
dist = dendropy.calculate.treecompare.symmetric_difference(tree1, tree2)
if normalized == 'F':
return dist
elif normalized == 'T':
max_RF = 2 * (len(taxa) - 2)
norm_dist = dist / max_RF
both = [dist, norm_dist]
return both
def main():
# Paths to reference and current tree to be compared
#pathRef = '../data/asymmetric_0.5/asymmetric_0.5.tree'
pathRef = '../data/symmetric_0.5/symmetric_0.5.tree'
pathCal = '../data/treeout.txt'
# Set the same taxon_set for all trees!
taxa = dendropy.TaxonSet()
refTree = dendropy.Tree.get_from_path(pathRef, schema="newick", taxon_set=taxa)
calTree = dendropy.Tree.get_from_path(pathCal, schema="newick", taxon_set=taxa)
# Open result-file
handle = open('../results/distances.txt', 'a')
# Compare the trees and append to file, differentiates noise reduced and normal trees
if sys.argv[1] == 'reduced':
handle.write(str(dendropy.treecalc.symmetric_difference(calTree, refTree)) + '\t')
else:
handle.write(str(dendropy.treecalc.symmetric_difference(calTree, refTree)) + '\n')
handle.close
def random_coal(self, nspecies=None, names=None):
if names and nspecies:
if not nspecies == len(names):
nspecies = len(names)
elif names and not nspecies:
nspecies = len(names)
elif not names:
if not nspecies:
nspecies = 16
names = taxonnames.names[:nspecies]
if nspecies > len(taxonnames.names):
names.extend(['Sp{0}'.format(i) for i in
range(len(taxonnames.names) + 1, nspecies
+ 1)])
taxon_set = dpy.TaxonSet(names)
tree = treesim.pure_kingman(taxon_set)
newick = '[&R] ' + tree.as_newick_string()
if not newick.endswith(';'):
newick += ';'
return Tree(newick)
def __init__(self,
work_queue,
result_split_dist_queue,
result_topology_hash_map_queue,
schema,
taxon_labels,
is_rooted,
ignore_node_ages,
calc_tree_probs,
weighted_trees,
tree_offset,
process_idx,
messenger,
messenger_lock,
log_frequency=1000):
multiprocessing.Process.__init__(self)
self.work_queue = work_queue
self.result_split_dist_queue = result_split_dist_queue
self.result_topology_hash_map_queue = result_topology_hash_map_queue
self.schema = schema
self.taxon_labels = list(taxon_labels)
self.taxon_set = dendropy.TaxonSet(self.taxon_labels)
self.split_distribution = treesplit.SplitDistribution(
taxon_set=self.taxon_set)
self.split_distribution.is_rooted = is_rooted
self.split_distribution.ignore_node_ages = ignore_node_ages
self.is_rooted = is_rooted
self.calc_tree_probs = calc_tree_probs
self.topology_counter = treesum.TopologyCounter()
self.weighted_trees = weighted_trees
self.tree_offset = tree_offset
self.process_idx = process_idx
self.messenger = messenger
self.messenger_lock = messenger_lock
self.log_frequency = log_frequency
self.kill_received = False