def test_wrapped_data(examples):
r = NexusReader.from_file(examples / 'example-wrapped-data.nex')
assert r.data.format['gap'] == '-'
assert len(r.data.taxa) == 3 and r.data.nchar == 471
sites = r.data.matrix[r.data.taxa[-1]]
assert len([s for s in sites if s == r.data.format['gap']]) == 5
if sys.version_info >= (3, 6):
assert len([s for s in sites if s == r.data.format['missing']]) == 85
def get_nexus_reader(thing):
if isinstance(thing, str):
return NexusReader.from_string(thing)
if isinstance(thing, pathlib.Path):
return NexusReader.from_file(thing)
if isinstance(thing, NexusWriter):
return NexusReader.from_string(thing.write())
assert isinstance(thing, NexusReader)
return thing
def make_indo_european_array():
n = NexusReader.from_file('IELex_Bouckaert2012.nex')
df = pd.DataFrame.from_dict(n.data.matrix, orient='index')
df = df.replace('?', np.nan)
# cheating at the moment by replacing '?' with 0
df = df.replace(np.nan, 0)
array = df.to_numpy()
array = np.ndarray.astype(array, dtype=np.float32)
missing_data_matrix = np.ones(array.shape, dtype=np.float32)
number_of_clusters = 7
features = 6280
samples = 103
return array, missing_data_matrix, samples, features, df
def main(args):
newick = ''
if args.newick:
newick = args.newick.readline()
elif args.nexus:
fn_nexus = args.nexus.name
nexus = NexusReader.from_file(fn_nexus)
newick = nexus.trees.trees[0].split('=')[1].strip()
edges = tree_to_adjacency_list(newick)
# print adjacency list to stdout
for v1, v2 in edges:
print('{},{}'.format(v1, v2))
def get_reader(args, many=False, required_blocks=None):
res = []
for f in (args.filename if many else [args.filename]):
if f is None:
res.append(NexusReader.from_string(sys.stdin.read()))
else:
res.append(NexusReader.from_file(f))
if required_blocks:
for nex in res:
for block in required_blocks:
if not getattr(nex, block, None):
raise ParserError(
colored(
'Nexus file {0} has no {1} block'.format(
nex.filename, block),
'red',
attrs=['bold'],
))
return res if many else res[0]
def cmd_makecldf(self, args):
args.writer.cldf.add_component('LanguageTable')
args.writer.cldf.add_table(
'trees.csv',
{
"name": 'ID',
"propertyUrl": "http://cldf.clld.org/v1.0/terms.rdf#id"
},
{
"name": 'Name',
"propertyUrl": "http://cldf.clld.org/v1.0/terms.rdf#name"
},
{
"name": "rooted",
"datatype": "boolean"
},
"Newick",
"Type",
"dplace_ID",
"source",
)
#
# FIXME: need to store the original tree ID and add the source!
#
t = args.writer.cldf.add_table(
'treelabels.csv',
{
"name": 'ID',
"propertyUrl": "http://cldf.clld.org/v1.0/terms.rdf#id"
},
{
"name": 'Name',
"propertyUrl": "http://cldf.clld.org/v1.0/terms.rdf#name"
},
{
"name":
"Language_ID",
"propertyUrl":
"http://cldf.clld.org/v1.0/terms.rdf#languageReference"
},
{
"name": "Tree_ID",
"separator": " "
},
)
t.add_foreign_key('Tree_ID', 'trees.csv', 'ID')
gcs = set()
treelabel_id = 0
tree_id = 0
for d in sorted(self.raw_dir.iterdir(), key=lambda p: p.stem):
if d.is_dir():
print(d.stem)
labels = {
l['taxon']: l['glottocode']
for l in reader(d.joinpath('taxa.csv'), dicts=True)
}
tree_ids = []
if d.joinpath('summary.trees').exists():
nx = NexusReader.from_file(d.joinpath('summary.trees'))
nx.trees.detranslate()
tree = nx.trees[0]
newick = loads(tree.newick_string, strip_comments=True)[0]
for n in newick.walk():
n.name = labels.get(n.name, n.name)
tree_id += 1
args.writer.objects['trees.csv'].append({
'ID':
str(tree_id),
'Name':
tree.name,
'rooted':
tree.rooted,
'Newick':
newick.newick,
"Type":
'summary',
"dplace_id":
d.stem,
})
tree_ids.append(tree_id)
if d.joinpath('posterior.trees').exists():
nx = NexusReader.from_file(d.joinpath('posterior.trees'))
nx.trees.detranslate()
for i, tree in enumerate(nx.trees, start=1):
newick = loads(tree.newick_string,
strip_comments=True)[0]
for n in newick.walk():
n.name = labels.get(n.name, n.name)
tree_id += 1
args.writer.objects['trees.csv'].append({
'ID':
str(tree_id),
'Name':
tree.name,
'rooted':
tree.rooted,
'Newick':
newick.newick,
"Type":
'sample',
"dplace_id":
d.stem,
})
tree_ids.append(tree_id)
for name, gc in sorted(labels.items()):
if gc:
gcs.add(gc)
treelabel_id += 1
args.writer.objects['treelabels.csv'].append({
'ID':
str(treelabel_id),
'Name':
name,
'Language_ID':
gc,
'Tree_ID': [str(i) for i in tree_ids],
})
for gc in sorted(gcs):
lang = args.glottolog.api.cached_languoids.get(gc)
if not lang:
args.log.warning('invalid glottocode: {0}'.format(gc))
args.writer.objects['LanguageTable'].append({
'ID':
gc,
'Name':
lang.name if lang else None,
'Latitude':
lang.latitude if lang else None,
'Longitude':
lang.longitude if lang else None,
'ISO639P3code':
lang.iso if lang else None,
'Glottocode':
gc,
})
def _make(fname):
return NexusReader.from_file(examples / fname)
