def write_newick_ott(out, ott, ott_id2children, root_ott_id, label_style,
prune_flags):
'''`out` is an output stream
`ott` is an OTT instance used for translating labels
`ott_id2children` is a dict mapping an OTT ID to the IDs of its children
`root_ott_id` is the root of the subtree to write.
`label_style` is a facet of OTULabelStyleEnum
`prune_flags` is a set strings (flags) or OTTFlagUnion instance or None
'''
if prune_flags is not None:
if not isinstance(prune_flags, OTTFlagUnion):
prune_flags = ott.convert_flag_string_set_to_union(prune_flags)
if ott.has_flag_set_key_intersection(root_ott_id, prune_flags):
return
else:
prune_flags = None
stack = [root_ott_id]
first_children = set(stack)
last_children = set(stack)
out.write('(')
while stack:
ott_id = stack.pop()
if isinstance(ott_id, tuple):
ott_id = ott_id[0]
else:
children = ott_id2children[ott_id]
if prune_flags is not None:
children = [
i for i in children
if not ott.has_flag_set_key_intersection(i, prune_flags)
]
if ott_id not in first_children:
out.write(',')
else:
first_children.remove(ott_id)
if bool(children):
out.write('(')
first_children.add(children[0])
last_children.add(children[-1])
stack.append(
(ott_id, )) # a tuple will signal exiting a node...
stack.extend([i for i in reversed(children)])
continue
n = ott.get_label(ott_id, label_style)
n = quote_newick_name(n)
out.write(n)
if ott_id in last_children:
out.write(')')
last_children.remove(ott_id)
out.write(';')
def write_newick_ott(out, ott, ott_id2children, root_ott_id, label_style, prune_flags):
'''`out` is an output stream
`ott` is an OTT instance used for translating labels
`ott_id2children` is a dict mapping an OTT ID to the IDs of its children
`root_ott_id` is the root of the subtree to write.
`label_style` is a facet of OTULabelStyleEnum
`prune_flags` is a set strings (flags) or OTTFlagUnion instance or None
'''
if prune_flags is not None:
if not isinstance(prune_flags, OTTFlagUnion):
prune_flags = ott.convert_flag_string_set_to_union(prune_flags)
if ott.has_flag_set_key_intersection(root_ott_id, prune_flags):
return
else:
prune_flags = None
stack = [root_ott_id]
first_children = set(stack)
last_children = set(stack)
out.write('(')
while stack:
ott_id = stack.pop()
if isinstance(ott_id, tuple):
ott_id = ott_id[0]
else:
children = ott_id2children[ott_id]
if prune_flags is not None:
children = [i for i in children if not ott.has_flag_set_key_intersection(i, prune_flags)]
if ott_id not in first_children:
out.write(',')
else:
first_children.remove(ott_id)
if bool(children):
out.write('(')
first_children.add(children[0])
last_children.add(children[-1])
stack.append((ott_id,)) # a tuple will signal exiting a node...
stack.extend([i for i in reversed(children)])
continue
n = ott.get_label(ott_id, label_style)
n = quote_newick_name(n)
out.write(n)
if ott_id in last_children:
out.write(')')
last_children.remove(ott_id)
out.write(';')
def write_newick_ott(out,
ott,
ott_id2children,
root_ott_id,
label_style,
prune_flags,
create_log_dict=False):
"""`out` is an output stream
`ott` is an OTT instance used for translating labels
`ott_id2children` is a dict mapping an OTT ID to the IDs of its children
`root_ott_id` is the root of the subtree to write.
`label_style` is a facet of OTULabelStyleEnum
`prune_flags` is a set strings (flags) or OTTFlagUnion instance or None
if `create_log_dict` is True, a dict will be returned that contains statistics
about the pruning.
"""
# create to_prune_fsi_set a set of flag set indices to prune...
if prune_flags:
flags_to_prune_list = list(prune_flags)
to_prune_fsi_set = ott.convert_flag_string_set_to_union(
flags_to_prune_list)
else:
flags_to_prune_list = []
to_prune_fsi_set = None
flags_to_prune_set = frozenset(flags_to_prune_list)
pfd = {}
log_dict = None
if create_log_dict:
log_dict = {
'version': ott.version,
'flags_to_prune': flags_to_prune_list
}
fsi_to_str_flag_set = {}
for k, v in dict(ott.flag_set_id_to_flag_set).items():
fsi_to_str_flag_set[k] = frozenset(list(v))
if to_prune_fsi_set:
pfd = {}
for f in to_prune_fsi_set.keys():
s = fsi_to_str_flag_set[f]
str_flag_intersection = flags_to_prune_set.intersection(s)
pfd[f] = list(str_flag_intersection)
pfd[f].sort()
# log_dict['prune_flags_d'] = d
# log_dict['pfd'] = pfd
pruned_dict = {}
num_tips = 0
num_pruned_anc_nodes = 0
num_nodes = 0
num_monotypic_nodes = 0
if to_prune_fsi_set and ott.has_flag_set_key_intersection(
root_ott_id, to_prune_fsi_set):
# entire taxonomy is pruned off
if log_dict is not None:
fsi = ott.get_flag_set_key(root_ott_id)
pruned_dict[fsi] = {'': [root_ott_id]}
num_pruned_anc_nodes += 1
else:
stack = [root_ott_id]
first_children = set(stack)
last_children = set()
while stack:
ott_id = stack.pop()
if isinstance(ott_id, tuple):
ott_id = ott_id[0]
else:
num_nodes += 1
children = ott_id2children[ott_id]
if to_prune_fsi_set is not None:
c = []
for child_id in children:
if ott.has_flag_set_key_intersection(
child_id, to_prune_fsi_set):
if log_dict is not None:
fsi = ott.get_flag_set_key(child_id)
fd = pruned_dict.get(fsi)
if fd is None:
pruned_dict[fsi] = {
'anc_ott_id_pruned': [child_id]
}
else:
fd['anc_ott_id_pruned'].append(child_id)
num_pruned_anc_nodes += 1
else:
c.append(child_id)
children = c
nc = len(children)
if nc < 2:
if nc == 1:
num_monotypic_nodes += 1
else:
num_tips += 1
if ott_id not in first_children:
out.write(',')
else:
first_children.remove(ott_id)
if bool(children):
out.write('(')
first_children.add(children[0])
last_children.add(children[-1])
stack.append(
(ott_id, )) # a tuple will signal exiting a node...
stack.extend([i for i in reversed(children)])
continue
n = ott.get_label(ott_id, label_style)
n = quote_newick_name(n)
out.write(n)
if ott_id in last_children:
out.write(')')
last_children.remove(ott_id)
out.write(';')
if create_log_dict:
log_dict['pruned'] = {}
for fsi, obj in pruned_dict.items():
f = pfd[fsi]
f.sort()
obj['flags_causing_prune'] = f
nk = ','.join(f)
log_dict['pruned'][nk] = obj
log_dict['num_tips'] = num_tips
log_dict['num_pruned_anc_nodes'] = num_pruned_anc_nodes
log_dict['num_nodes'] = num_nodes
log_dict['num_non_leaf_nodes'] = num_nodes - num_tips
log_dict[
'num_non_leaf_nodes_with_multiple_children'] = num_nodes - num_tips - num_monotypic_nodes
log_dict['num_monotypic_nodes'] = num_monotypic_nodes
return log_dict
# read pruned labelled synthesis tree
ottpattern = re.compile(r"([(,)])(ott)(\d+)")
mrcapattern = re.compile(r"([(,)])mrcaott(\d+)ott(\d+)")
pos = 0
ott = OTT(ott_dir=args.ott_dir)
# load up the OTT dictionary...
d = ott.ott_id_to_names
outfile = codecs.open('ottnamelabelledtree.tre', 'w', encoding='utf-8')
with open(args.newick_file, 'r') as f:
newick = f.read()
for m in re.finditer(ottpattern, newick):
#print m.group(1),m.group(2),m.group(3)
ottid = int(m.group(3))
ottresults = d[ottid]
ottname = ottresults
if isinstance(ottresults, tuple):
ottname = ottresults[0]
print m.group(3), ottname
skippedchars = newick[pos:m.start()]
outfile.write(skippedchars)
outfile.write(m.group(1))
if args.keep_ottids:
label = quote_newick_name('{} ott{}'.format(ottname, ottid))
else:
label = quote_newick_name('{}'.format(ottname))
outfile.write(label)
pos = m.end()
outfile.write(newick[pos:])
f.close()
outfile.close()
def write_newick_ott(out,
ott,
ott_id2children,
root_ott_id,
label_style,
prune_flags,
create_log_dict=False):
"""`out` is an output stream
`ott` is an OTT instance used for translating labels
`ott_id2children` is a dict mapping an OTT ID to the IDs of its children
`root_ott_id` is the root of the subtree to write.
`label_style` is a facet of OTULabelStyleEnum
`prune_flags` is a set strings (flags) or OTTFlagUnion instance or None
if `create_log_dict` is True, a dict will be returned that contains statistics
about the pruning.
"""
# create to_prune_fsi_set a set of flag set indices to prune...
if prune_flags:
flags_to_prune_list = list(prune_flags)
to_prune_fsi_set = ott.convert_flag_string_set_to_union(flags_to_prune_list)
else:
flags_to_prune_list = []
to_prune_fsi_set = None
flags_to_prune_set = frozenset(flags_to_prune_list)
pfd = {}
log_dict = None
if create_log_dict:
log_dict = {'version': ott.version, 'flags_to_prune': flags_to_prune_list}
fsi_to_str_flag_set = {}
for k, v in dict(ott.flag_set_id_to_flag_set).items():
fsi_to_str_flag_set[k] = frozenset(list(v))
if to_prune_fsi_set:
pfd = {}
for f in to_prune_fsi_set.keys():
s = fsi_to_str_flag_set[f]
str_flag_intersection = flags_to_prune_set.intersection(s)
pfd[f] = list(str_flag_intersection)
pfd[f].sort()
# log_dict['prune_flags_d'] = d
# log_dict['pfd'] = pfd
pruned_dict = {}
num_tips = 0
num_pruned_anc_nodes = 0
num_nodes = 0
num_monotypic_nodes = 0
if to_prune_fsi_set and ott.has_flag_set_key_intersection(root_ott_id, to_prune_fsi_set):
# entire taxonomy is pruned off
if log_dict is not None:
fsi = ott.get_flag_set_key(root_ott_id)
pruned_dict[fsi] = {'': [root_ott_id]}
num_pruned_anc_nodes += 1
else:
stack = [root_ott_id]
first_children = set(stack)
last_children = set()
while stack:
ott_id = stack.pop()
if isinstance(ott_id, tuple):
ott_id = ott_id[0]
else:
num_nodes += 1
children = ott_id2children[ott_id]
if to_prune_fsi_set is not None:
c = []
for child_id in children:
if ott.has_flag_set_key_intersection(child_id, to_prune_fsi_set):
if log_dict is not None:
fsi = ott.get_flag_set_key(child_id)
fd = pruned_dict.get(fsi)
if fd is None:
pruned_dict[fsi] = {'anc_ott_id_pruned': [child_id]}
else:
fd['anc_ott_id_pruned'].append(child_id)
num_pruned_anc_nodes += 1
else:
c.append(child_id)
children = c
nc = len(children)
if nc < 2:
if nc == 1:
num_monotypic_nodes += 1
else:
num_tips += 1
if ott_id not in first_children:
out.write(',')
else:
first_children.remove(ott_id)
if bool(children):
out.write('(')
first_children.add(children[0])
last_children.add(children[-1])
stack.append((ott_id,)) # a tuple will signal exiting a node...
stack.extend([i for i in reversed(children)])
continue
n = ott.get_label(ott_id, label_style)
n = quote_newick_name(n)
out.write(n)
if ott_id in last_children:
out.write(')')
last_children.remove(ott_id)
out.write(';')
if create_log_dict:
log_dict['pruned'] = {}
for fsi, obj in pruned_dict.items():
f = pfd[fsi]
f.sort()
obj['flags_causing_prune'] = f
nk = ','.join(f)
log_dict['pruned'][nk] = obj
log_dict['num_tips'] = num_tips
log_dict['num_pruned_anc_nodes'] = num_pruned_anc_nodes
log_dict['num_nodes'] = num_nodes
log_dict['num_non_leaf_nodes'] = num_nodes - num_tips
log_dict['num_non_leaf_nodes_with_multiple_children'] = num_nodes - num_tips - num_monotypic_nodes
log_dict['num_monotypic_nodes'] = num_monotypic_nodes
return log_dict