def testUnnormalized(self):
for i in range(0xFF):
s = bitprocessing.int_as_bitstring(i, 8, ".", "*")[::-1]
r = paup.PaupService.bipartition_groups_to_split_bitmask(
s, normalized=False)
self.assertEqual(r, i, "%s => %s => %s" \
% (bitprocessing.int_as_bitstring(i, 8), s, bitprocessing.int_as_bitstring(r, 8)))
def testNormalized1(self):
for i in range(0xFF):
s = bitprocessing.int_as_bitstring(i, 8, ".", "*")[::-1]
r = paup.PaupService.bipartition_groups_to_split_bitmask(s, normalized=True)
normalized = Bipartition.normalize_bitmask(i, 0xFF, 1)
self.assertEqual(r, normalized, "%s => %s => %s" \
% (bitprocessing.int_as_bitstring(i, 8), s, bitprocessing.int_as_bitstring(normalized, 8)))
def testNormalized1(self):
for i in range(0xFF):
s = bitprocessing.int_as_bitstring(i, 8, ".", "*")[::-1]
r = paup.PaupService.bipartition_groups_to_split_bitmask(
s, normalized=True)
normalized = Bipartition.normalize_bitmask(i, 0xFF, 1)
self.assertEqual(r, normalized, "%s => %s => %s" \
% (bitprocessing.int_as_bitstring(i, 8), s, bitprocessing.int_as_bitstring(normalized, 8)))
def split_mapper(tree, taxon_set):
split_dict={}
for node in tree.postorder_node_iter():
if node.parent_node is None:
node.value = 1.0
else:
taxlist=[]
tax = taxon_set.bitmask_taxa_list(node.edge.split_bitmask)
for i in tax:
taxlist.append(i.label)
split_dict[bitprocessing.int_as_bitstring(node.edge.split_bitmask, length=len(taxon_set))] = taxlist
return split_dict
def split_freq(tree,taxon_set, split_occ_dict): #for each node for node in tree.postorder_node_iter(): #if not root if node.parent_node is None: node.value = 1.0 else: #split=dendropy.treesplit.split_as_string(node.edge.split_bitmask, width=len(taxon_set)) #deprecated split=bitprocessing.int_as_bitstring(node.edge.split_bitmask, length=len(taxon_set)) print(split) if split not in split_occ_dict.keys(): split_occ_dict[split] = [1] else: split_occ_dict[split].append(1) return split_occ_dict
def split_as_string_rev(split_mask, width, symbol1='.', symbol2='*'):
deprecate.dendropy_deprecation_warning(
preamble="Deprecated since DendroPy 4: 'dendropy.treesplit.split_as_string_rev()'.",
old_construct="from dendropy import treesplit\nd = treesplit.split_as_string(...)",
new_construct="""\
# if using a bipartition
d = bipartition.split_as_bitstring(...)[::-1]
d = bipartition.leafset_as_bitstring(...)[::-1]
# if a "raw" bitmask
from dendropy.utility import bitprocessing
d = bitprocessing.int_as_bitstring(..., reverse=True)""")
return bitprocessing.int_as_bitstring(
mask=split_mask,
length=width,
symbol0=symbol0,
symbol1=symbol1,
reverse=True)
def split_as_string(split_mask, width, symbol1=None, symbol2=None):
deprecate.dendropy_deprecation_warning(
preamble="Deprecated since DendroPy 4: 'dendropy.treesplit.split_as_string()'.",
old_construct="from dendropy import treesplit\nd = treesplit.split_as_string(...)",
new_construct="""\
# if using a bipartition
d = bipartition.split_as_bitstring(...)
d = bipartition.leafset_as_bitstring(...)
# if a "raw" bitmask
from dendropy.utility import bitprocessing
d = bitprocessing.int_as_bitstring(...)""")
return bitprocessing.int_as_bitstring(
n=split_mask,
length=width,
symbol0=symbol1,
symbol1=symbol2,
reverse=False)
def split_as_string(split_mask, width, symbol1=None, symbol2=None):
deprecate.dendropy_deprecation_warning(
preamble=
"Deprecated since DendroPy 4: 'dendropy.treesplit.split_as_string()'.",
old_construct=
"from dendropy import treesplit\nd = treesplit.split_as_string(...)",
new_construct="""\
# if using a bipartition
d = bipartition.split_as_bitstring(...)
d = bipartition.leafset_as_bitstring(...)
# if a "raw" bitmask
from dendropy.utility import bitprocessing
d = bitprocessing.int_as_bitstring(...)""")
return bitprocessing.int_as_bitstring(n=split_mask,
length=width,
symbol0=symbol1,
symbol1=symbol2,
reverse=False)
def split_as_string_rev(split_mask, width, symbol1='.', symbol2='*'):
deprecate.dendropy_deprecation_warning(
preamble=
"Deprecated since DendroPy 4: 'dendropy.treesplit.split_as_string_rev()'.",
old_construct=
"from dendropy import treesplit\nd = treesplit.split_as_string(...)",
new_construct="""\
# if using a bipartition
d = bipartition.split_as_bitstring(...)[::-1]
d = bipartition.leafset_as_bitstring(...)[::-1]
# if a "raw" bitmask
from dendropy.utility import bitprocessing
d = bitprocessing.int_as_bitstring(..., reverse=True)""")
return bitprocessing.int_as_bitstring(mask=split_mask,
length=width,
symbol0=symbol0,
symbol1=symbol1,
reverse=True)
def getBitStringLabels(tree, l): tree2 = tree.clone(depth=1) for node in tree2.postorder_node_iter(): lab = "\"" + str(bitprocessing.int_as_bitstring(node.edge.split_bitmask, length=l)) + "\"" node.label=lab return tree2
def main():
params = parseArgs()
print("\n####################anomaly_finder.py######################\n")
#read trees
taxon_set = dendropy.TaxonNamespace()
print("Reading primary tree...", params.tree)
mrc = dendropy.Tree.get_from_path(src=params.tree, schema=params.ftype, taxon_namespace = taxon_set)
if params.btrees:
print("Reading bootstrap trees...", params.btrees)
trees = dendropy.TreeList.get_from_path(params.btrees, params.ftype, taxon_namespace = taxon_set)
else:
print("No bootstrap trees provided.")
#initialize some dictionaries
master_dict=dict()
mrc_dict=dict()
split_dict=dict()
split_occ_dict=dict()
#if bootstrap trees provided, get results for them
#master_dict:
#key = pair of edges
#value = list of anomaly zone results (1=True; 0=False)
if params.btrees:
print("Finding anomalous nodes in bootsteap trees...")
for sp_tree in trees:
#dendropy.treesplit.encode_splits(sp_tree) #deprecated
sp_tree.encode_bipartitions()
#split_bitmasks = sp_tree.split_bitmask_edge_map.keys()
master_dict = get_nodes(sp_tree, master_dict)
split_occ_dict=split_freq(sp_tree, taxon_set, split_occ_dict)
#sys.exit(0)
#get results for main tree
print("Finding anomalous nodes in primary tree...")
mrc.encode_bipartitions()
mrc_dict=get_nodes(mrc, mrc_dict) #get AZ results for primary tree
split_dict=split_mapper(mrc, taxon_set) #match a list of taxa to the split pattern from the bitmask for the whole tree
print("\n######################ENCODING#############################\n")
print("Writing results with splits encoded as:")
labelTree = getBitStringLabels(mrc, len(taxon_set))
labelTree.print_plot(show_internal_node_labels=True)
print("(If tree is large, import the following into FigTree)\n")
print(labelTree.as_string(schema="newick"))
#if bootstrap trees provided:
#for each internal edge pair:
print("\n######################RESULTS##############################\n")
print("Showing anomalous nodes only:\n")
if params.btrees:
print("ParentEdge\tDescendantEdge\tProp.Anomalous")
seen=False
for k,v in master_dict.items():
taxlab1=list()
taxlab2=list()
#print(k[0])
#if edge pair has results for main tree
if k in mrc_dict.keys():
#print("k:",k, "v:",v)
#bitstring for parent node
parentLabel = bitprocessing.int_as_bitstring(k[0], length=len(taxon_set))
#print(parentLabel)
#list of descendent taxa
parentList = taxon_set.bitmask_taxa_list(k[0])
#print(parentList)
#label for descendant edge
descendantLabel = bitprocessing.int_as_bitstring(k[1], length=len(taxon_set))
#list of descendant taxa
descendantList = taxon_set.bitmask_taxa_list(k[1])
#print(descendantList)
for i in parentList:
taxlab1.append(i.label)
for j in descendantList:
taxlab2.append(j.label)
if sum(v) > 0:
seen=True
print(parentLabel, descendantLabel,sum(v)/len(v), sep="\t")
if seen==False:
print("No anomalous nodes detected.")
else:
print("ParentEdge\tDescendantEdge")
seen=False
for k,v in mrc_dict.items():
taxlab1=list()
taxlab2=list()
anom=False
#print("k:",k, "v:",v)
#bitstring for parent node
parentLabel = bitprocessing.int_as_bitstring(k[0], length=len(taxon_set))
#print(parentLabel)
#list of descendent taxa
parentList = taxon_set.bitmask_taxa_list(k[0])
#print(parentList)
#label for descendant edge
descendantLabel = bitprocessing.int_as_bitstring(k[1], length=len(taxon_set))
#list of descendant taxa
descendantList = taxon_set.bitmask_taxa_list(k[1])
#print(descendantList)
for i in parentList:
taxlab1.append(i.label)
for j in descendantList:
taxlab2.append(j.label)
if sum(v) > 0:
seen=True
print(parentLabel, descendantLabel, sep="\t")
if seen==False:
print("No anomalous nodes detected.")
print("\n########################OUTPUT############################\n")
if params.btrees:
print("Tree with anomalous pairs annotated as \"AZ Event : Prop. BS\"")
outTreeBS = getAnomalyLabelsBS(mrc, master_dict)
outTreeBS.print_plot(show_internal_node_labels=True)
print("(If tree is large, import the following into FigTree)\n")
print(outTreeBS.as_string(schema="newick"))
with open(params.out, 'w') as fh:
l=outTreeBS.as_string(schema="newick")
l=l+"\n"
fh.write(l)
fh.close()
else:
print("Tree with anomalous edge pairs annotated (AZ pairs numbered)")
outTree = getAnomalyLabels(mrc, mrc_dict)
outTree.print_plot(show_internal_node_labels=True)
print("(If tree is large, import the following into FigTree)\n")
print(outTree.as_string(schema="newick"))
with open(params.out, 'w') as fh:
l=outTree.as_string(schema="newick")
l=l+"\n"
fh.write(l)
fh.close()
print("\n########################DONE!############################\n\n")
def testUnnormalized(self):
for i in range(0xFF):
s = bitprocessing.int_as_bitstring(i, 8, ".", "*")[::-1]
r = paup.PaupService.bipartition_groups_to_split_bitmask(s, normalized=False)
self.assertEqual(r, i, "%s => %s => %s" \
% (bitprocessing.int_as_bitstring(i, 8), s, bitprocessing.int_as_bitstring(r, 8)))
