def read_from_tree(self, tree, warnings=True):
"""Obtain the taxonomy for each extant taxa as specified by internal tree labels.
Parameters
----------
tree : str or dendropy.Tree
Filename of newick tree or dendropy tree object.
Returns
-------
dict : d[unique_id] -> [d__<taxon>, ..., s__<taxon>]
Taxa indexed by unique ids.
"""
if isinstance(tree, str):
tree = dendropy.Tree.get_from_path(tree,
schema='newick',
rooting="force-rooted",
preserve_underscores=True)
taxonomy = {}
for leaf in tree.leaf_node_iter():
taxa = []
node = leaf.parent_node
while node:
if node.label:
taxa_str = node.label
if ':' in taxa_str:
taxa_str = taxa_str.split(':')[1]
if not is_float(taxa_str):
if taxa_str[-1] == ';':
taxa_str = taxa_str[:-1]
# check for concatenated ranks of the form: p__Crenarchaeota__c__Thermoprotei
for prefix in Taxonomy.rank_prefixes:
split_str = '__' + prefix
if split_str in taxa_str:
taxa_str = taxa_str.replace(split_str, ';' + prefix)
# appears to be an internal label and not simply a support value
taxa = [x.strip() for x in taxa_str.split(';')] + taxa
node = node.parent_node
if warnings and len(taxa) > 7:
self.logger.warning('Invalid taxonomy string read from tree for taxon %s: %s' % (leaf.taxon.label, taxa))
#sys.exit(-1)
# check if genus name should be appended to species label
if len(taxa) == 7:
genus = taxa[5][3:]
species = taxa[6][3:]
if genus not in species:
taxa[6] = 's__' + genus + ' ' + species
taxa = self.fill_trailing_ranks(taxa)
taxonomy[leaf.taxon.label] = taxa
return taxonomy
def read_from_tree(self, tree):
"""Obtain the taxonomy for each extant taxa as specified by internal tree labels.
Parameters
----------
tree : str or dendropy.Tree
Filename of newick tree or dendropy tree object.
Returns
-------
dict : d[unique_id] -> [d__<taxon>, ..., s__<taxon>]
Taxa indexed by unique ids.
"""
if isinstance(tree, basestring):
tree = dendropy.Tree.get_from_path(tree, schema='newick', rooting="force-rooted", preserve_underscores=True)
taxonomy = {}
for leaf in tree.leaf_node_iter():
taxa = []
node = leaf.parent_node
while node:
if node.label:
taxa_str = node.label
if ':' in taxa_str:
taxa_str = taxa_str.split(':')[1]
if not is_float(taxa_str):
# appears to be an internal label and not simply a support value
taxa = [x.strip() for x in taxa_str.split(';')] + taxa
node = node.parent_node
if len(taxa) > 7:
self.logger.error('Invalid taxonomy string read from tree for taxon %s: %s' % (leaf.taxon.label, taxa))
sys.exit(-1)
# check if genus name should be appended to species label
if len(taxa) == 7:
genus = taxa[5][3:]
species = taxa[6][3:]
if genus not in species:
taxa[6] = 's__' + genus + ' ' + species
taxa = self.fill_missing_ranks(taxa)
taxonomy[leaf.taxon.label] = taxa
return taxonomy
def parse_label(label):
"""Parse a Newick label which may contain a support value, taxon, and/or auxiliary information.
Parameters
----------
label : str
Internal label in a Newick tree.
Returns
-------
float
Support value specified by label, or None
str
Taxon specified by label, or None
str
Auxiliary information, on None
"""
support = None
taxon = None
auxiliary_info = None
if label:
label = label.strip()
if '|' in label:
label, auxiliary_info = label.split('|')
if ':' in label:
support, taxon = label.split(':')
support = float(support)
else:
if is_float(label):
support = float(label)
elif label != '':
taxon = label
return support, taxon, auxiliary_info
def parse_label(label):
"""Parse a Newick label which may contain a support value, taxon, and/or auxiliary information.
Parameters
----------
label : str
Internal label in a Newick tree.
Returns
-------
float
Support value specified by label, or None
str
Taxon specified by label, or None
str
Auxiliary information, on None
"""
support = None
taxon = None
auxiliary_info = None
if label:
label = label.strip()
if "|" in label:
label, auxiliary_info = label.split("|")
if ":" in label:
support, taxon = label.split(":")
support = float(support)
else:
if is_float(label):
support = float(label)
elif label != "":
taxon = label
return support, taxon, auxiliary_info
def _clade_pd(self, tree, ingroup_count, outgroup_count):
"""Calculate PD for named clades."""
pd = {}
for node in tree.preorder_node_iter():
if not node.label:
continue
taxon = None
if ':' in node.label:
_support, taxon = node.label.split(':')
else:
if not is_float(node.label):
taxon = node.label
if taxon:
taxon_pd = 0
taxon_count = 0
in_taxon_pd = 0
in_taxon_count = 0
in_taxon_derep = 0
out_taxon_pd = 0
out_taxon_count = 0
out_taxon_derep = 0
for nn in node.postorder_iter():
if nn == node:
continue
# check if group contains taxa from
# the ingroup and/or outgroup
ingroup_leaves = False
outgroup_leaves = False
for leaf in nn.leaf_iter():
genome_id = leaf.taxon.label
if genome_id in ingroup_count:
ingroup_leaves = True
if genome_id in outgroup_count:
outgroup_leaves = True
if ingroup_leaves:
in_taxon_pd += nn.edge.length
if outgroup_leaves:
out_taxon_pd += nn.edge.length
if nn.is_leaf():
genome_id = nn.taxon.label
in_taxon_count += ingroup_count.get(genome_id, 0)
if genome_id in ingroup_count:
in_taxon_derep += 1
out_taxon_count += outgroup_count.get(genome_id, 0)
if genome_id in outgroup_count:
out_taxon_derep += 1
taxon_pd += nn.edge.length
if taxon == 'd__Archaea':
print taxon_count, in_taxon_count, out_taxon_count, in_taxon_pd, out_taxon_pd
pd[taxon] = [taxon_pd, in_taxon_pd, in_taxon_count, in_taxon_derep, out_taxon_pd, out_taxon_count, out_taxon_derep]
return pd
def read_viral_taxonomy_from_tree(tree):
"""Obtain the taxonomy for each extant taxa as specified by internal tree labels.
Parameters
----------
tree : str or dendropy.Tree
Filename of newick tree or dendropy tree object.
Returns
-------
dict : d[unique_id] -> [d__<taxon>, ..., s__<taxon>]
Taxa indexed by unique ids.
"""
if isinstance(tree, str):
tree = dendropy.Tree.get_from_path(tree,
schema='newick',
rooting="force-rooted",
preserve_underscores=True)
taxonomy = {}
for leaf in tree.leaf_node_iter():
taxa = []
node = leaf.parent_node
while node:
if node.label:
taxa_str = node.label
if ':' in taxa_str:
taxa_str = taxa_str.split(':')[1]
if not is_float(taxa_str):
if taxa_str[-1] == ';':
taxa_str = taxa_str[:-1]
# appears to be an internal label and not simply a support value
taxa = [x.strip() for x in taxa_str.split(';')] + taxa
node = node.parent_node
if len(taxa) > 7:
logger = logging.getLogger()
logger.error(
'Invalid taxonomy string read from tree for taxon %s: %s' %
(leaf.taxon.label, ';'.join(taxa)))
sys.exit(1)
for taxon in taxa:
prefix = taxon[0:3]
if prefix not in VIRAL_PREFIX_TRANSLATION:
print('Unrecognized viral prefix for {}: {}'.format(
taxon, prefix))
sys.exit(1)
# fill missing ranks
try:
last_rank = VIRAL_RANK_PREFIXES.index(taxa[-1][0:3])
except:
logger = logging.getLogger()
logger.error('Taxon {} is missing rank prefix: {}'.format(
leaf.taxon.label, ';'.join(taxa)))
sys.exit(1)
for i in range(last_rank + 1, len(VIRAL_RANK_PREFIXES)):
taxa.append(VIRAL_RANK_PREFIXES[i])
taxonomy[leaf.taxon.label] = taxa
return taxonomy
def _clade_pd(self, tree, ingroup_count, outgroup_count):
"""Calculate PD for named clades."""
pd = {}
for node in tree.preorder_node_iter():
if not node.label:
continue
taxon = None
if ':' in node.label:
_support, taxon = node.label.split(':')
else:
if not is_float(node.label):
taxon = node.label
if taxon:
taxon_pd = 0
taxon_count = 0
in_taxon_pd = 0
in_taxon_count = 0
in_taxon_derep = 0
out_taxon_pd = 0
out_taxon_count = 0
out_taxon_derep = 0
for nn in node.postorder_iter():
if nn == node:
continue
# check if group contains taxa from
# the ingroup and/or outgroup
ingroup_leaves = False
outgroup_leaves = False
for leaf in nn.leaf_iter():
genome_id = leaf.taxon.label
if genome_id in ingroup_count:
ingroup_leaves = True
if genome_id in outgroup_count:
outgroup_leaves = True
if ingroup_leaves:
in_taxon_pd += nn.edge.length
if outgroup_leaves:
out_taxon_pd += nn.edge.length
if nn.is_leaf():
genome_id = nn.taxon.label
in_taxon_count += ingroup_count.get(genome_id, 0)
if genome_id in ingroup_count:
in_taxon_derep += 1
out_taxon_count += outgroup_count.get(genome_id, 0)
if genome_id in outgroup_count:
out_taxon_derep += 1
taxon_pd += nn.edge.length
if taxon == 'd__Archaea':
print taxon_count, in_taxon_count, out_taxon_count, in_taxon_pd, out_taxon_pd
pd[taxon] = [
taxon_pd, in_taxon_pd, in_taxon_count, in_taxon_derep,
out_taxon_pd, out_taxon_count, out_taxon_derep
]
return pd