def parse_newick_to_tree(filename):
with io.open(filename, encoding='ascii') as fp:
trees = newick.load(fp)
tree_root = TreeNode(trees[0].name, [])
for child in create_children(trees[0].descendants):
tree_root.addChild(child)
return tree_root
def species_list_from_tree(fname):
species = []
from newick import load
tree = load(file(fname))
for node in tree[0].walk():
if node.name:
species.append(node.name)
return species
def parse(file_name):
with io.open(file_name, encoding='utf8') as fp:
loaded_trees = load(fp)
trees = []
for t in loaded_trees:
tree = Tree(leaves=t.get_leaves())
load_tree(t, tree)
trees.append(tree)
return trees
def read_tree(filename):
"""Read tree into memory.
NOTE - The tree file must be in NEWICK format
Arguemnts:
filename - the location of the tree file.
Requires:
newick, io.
"""
with io.open(filename, encoding='utf8') as fp:
tree = newick.load(fp)
return tree[0]
args = parser.parse_args()
if args.semantic_network:
related_concepts = networkx.parse_gml(args.semantic_network)
else:
concept2field = defaultdict(set)
for c in range(args.concepts):
concept2field[random.randint(0, args.fields - 1)].add(c)
related_concepts = {}
for field in concept2field.values():
for concept in field:
related_concepts[concept] = field - {concept}
i = 0
for _, tree_file in enumerate(args.trees):
for tree in newick.load(tree_file):
i += 1
phy = Phylogeny(related_concepts,
basic=[],
tree=tree,
scale=args.scale)
phy.simulate(p_loss=args.p_loss, p_gain=args.p_gain, p_new=args.p_new)
# "basic" is the number of words we afterwards use to to infer
# phylogeny with neighbor-joining
dataframe, columns = phy.collect_word_list(
Language.vocabulary,
collect_tips_only=not args.sample_internal_nodes)
if args.lodict is None:
lodict = {}
else:
lodict = pickle.load(args.lodict)
data = pandas.io.parsers.read_csv(
args.input,
sep="\t",
na_values=[""],
keep_default_na=False,
encoding='utf-8',
index_col=["English", "Language_ID", "Tokens"])
if args.guide_tree:
tree = newick.load(args.guide_tree)[0]
else:
# Calculate an UPGMA tree or something
by_language = data.groupby("Language_ID")
languages = []
distance_matrix = numpy.zeros((len(by_language), len(by_language)))
for (l1, (language1, data1)), (l2, (language2,
data2)) in (itertools.combinations(
enumerate(by_language), 2)):
if language1 not in languages:
languages.append(language1)
print(language1, language2)
c = 0
shared_vocab = 0
for concept in set(data1["Feature_ID"]) | set(data2["Feature_ID"]):
c1 = set(data1["Cognate Set"][data1["Feature_ID"] == concept])
node.length += height-ht
return root.length + height, root
return root.length, root
if __name__ == "__main__":
parser = argparse.ArgumentParser()
parser.add_argument("trees", type=argparse.FileType('r'))
parser.add_argument("--output", "-o",
type=argparse.FileType('w'),
default=sys.stdout,
help="Output filename")
parser.add_argument(
"--mean", action="store_const", dest="acc",
const=lambda x: sum(x)/len(x), default=lambda x: sum(x)/len(x),
help="Use arithmetic mean to calculate new branch height")
parser.add_argument(
"--rms", action="store_const", dest="acc",
const=lambda x: (sum(map(lambda xi: xi*xi, x))/len(x))**0.5,
help="Use root mean square to calculate new branch height")
parser.add_argument(
"--max", action="store_const", dest="acc",
const=max,
help="Use maximum to calculate new branch height")
args = parser.parse_args()
trees = newick.load(args.trees)
for tree in trees:
normalize_tree(tree, args.acc)
newick.dump(trees, args.output)
s_data = random_observed_data(tree, models)
s_names, s_classes = data_pattern(s_data)
if classes != s_classes:
if verbose:
print("Not generating the right data: {:}".format(tree.newick))
yield None
continue
for old, new in zip(s_names, names):
tree.get_node(old).name = new
yield tree
if __name__ == "__main__":
try:
with open("true.tree") as treefile:
true_tree = newick.load(treefile)[0]
except (ValueError, FileNotFoundError):
true_tree = random_tree()
with open("true.tree", "w") as treefile:
print(true_tree.newick, file=treefile, end=";\n", flush=True)
try:
with open("data.json") as datafile:
true_data = json.load(datafile)
except (json.JSONDecodeError, FileNotFoundError):
true_data = random_observed_data(true_tree)
with open("data.json", "w") as datafile:
json.dump(true_data, datafile, indent=2)
print(true_tree.newick)
print(true_data)