def parse_tree_tips(tree_dir):
tips = []
tip_to_tree = {}
for fn in os.listdir(tree_dir):
if fn.endswith(".tree"):
tree_name = fn.split(".")[0]
tree = bt.loadNewick(tree_dir + "/" + fn, absoluteTime=False)
for k in tree.Objects:
if k.branchType == 'leaf' and "inserted" not in k.name:
tips.append(k.name)
tip_to_tree[k.name] = tree_name
elif fn.endswith(".txt"):
tree_name = fn.split(".")[0]
with open(tree_dir + "/" + fn) as f:
for l in f:
tip_string = l.strip("\n").split("\t")[1]
tip_list = tip_string.split(",")
tips.extend(tip_list)
for i in tip_list:
tip_to_tree[i] = tree_name
return tips, tip_to_tree
def find_tallest_tree(input_dir):
tree_heights = []
for r, d, f in os.walk(input_dir):
for fn in f:
if fn.endswith(".tree"):
num_taxa = 0
intro_name = ""
with open(r + '/' + fn, "r") as f:
for l in f:
l = l.rstrip("\n")
if l.startswith(" Dimensions NTax="):
num_taxa = int(l.rstrip(";").split("=")[1])
intro_name = fn.rstrip(".tree")
if num_taxa > 1:
tree_file = os.path.join(r, fn)
tree = bt.loadNewick(tree_file, absoluteTime=False)
tips = []
for k in tree.Objects:
if k.branchType == 'leaf':
tips.append(k.name)
tree_heights.append(tree.treeHeight)
max_height = sorted(tree_heights, reverse=True)[0]
return max_height
def relabel_tips(treeFile, outTree):
myTree = bt.loadNewick(treeFile, absoluteTime=False)
#myTree.setAbsoluteTime(2019.00) # need to set this to time of last sampled tip
myTree.traverse_tree() ## required to set heights
myTree.treeStats() ## report stats about tree
for k in myTree.Objects: ## iterate over a flat list of branches
if k.branchType == 'leaf':
curr_name = k.numName
if 'type' in k.traits:
ttype = k.traits['type']
else:
ttype = 'Unknown' #Assign tip unknown type if not given
if 'time' in k.traits:
time = k.traits['time']
else:
time = 'Unknown'
k.numName = curr_name + '_' + str(time) + '_' + ttype
print(k.numName)
newick = myTree.toString(traits=[], numName=True, nexus=False)
newick = newick.replace('\'', '')
tfile = open(outTree, "w")
tfile.write(newick)
tfile.close()
def test_newick(self):
tree = bt.loadNewick('./tests/data/zika.nwk')
expected_num_nodes = 564
assert len(tree.Objects) == expected_num_nodes, 'Newick tree does not contain correct number of nodes. Expected: {}. Observed: {}'.format(expected_num_nodes, len(tree.Objects))
max_height = round(max([i.height for i in tree.Objects]), 4)
expected_height = 0.0058
assert max_height == expected_height, 'Newick tree height is not correct. Expected: {}. Observed: {}'.format(expected_height, max_height)
def find_tallest_tree(input_dir):
tree_heights = []
for r, d, f in os.walk(input_dir):
for fn in f:
if fn.endswith(".tree"):
tree_file = os.path.join(r, fn)
tree = bt.loadNewick(tree_file, absoluteTime=False)
tips = []
for k in tree.Objects:
if k.branchType == 'leaf':
tips.append(k.name)
tree_heights.append(tree.treeHeight)
max_height = sorted(tree_heights, reverse=True)[0]
return max_height
def relabel_tips(treeFile, outTree):
myTree = bt.loadNewick(treeFile, absoluteTime=False)
#myTree.setAbsoluteTime(2019.00) # need to set this to time of last sampled tip
myTree.traverse_tree() ## required to set heights
myTree.treeStats() ## report stats about tree
for k in myTree.Objects: ## iterate over a flat list of branches
if k.branchType == 'leaf':
curr_name = k.numName
curr_name = curr_name.replace('|', '_') + '_Il'
curr_name = curr_name.replace('2020_EPI_ISL_', '')
k.numName = curr_name
print(k.numName)
newick = myTree.toString(traits=[], numName=True, nexus=False)
newick = newick.replace('\'', '')
tfile = open(outTree, "w")
tfile.write(newick)
tfile.close()
def make_all_of_the_trees(input_dir,
tree_name_stem,
taxon_dict,
query_dict,
desired_fields,
custom_tip_labels,
graphic_dict,
min_uk_taxa=3):
tallest_height = find_tallest_tree(input_dir)
too_tall_trees = []
colour_dict_dict = defaultdict(dict)
overall_df_dict = defaultdict(dict)
overall_tree_count = 0
lst = sort_trees_index(input_dir)
for trait, colour_scheme in graphic_dict.items():
colour_dict = find_colour_dict(query_dict, trait, colour_scheme)
colour_dict_dict[trait] = colour_dict
for fn in lst:
lineage = fn
treename = f"{tree_name_stem}_{fn}"
treefile = f"{tree_name_stem}_{fn}.tree"
nodefile = f"{tree_name_stem}_{fn}"
num_taxa = 0
intro_name = ""
with open(input_dir + "/" + treefile, "r") as f:
for l in f:
l = l.rstrip("\n")
if l.startswith(" Dimensions NTax="):
num_taxa = int(l.rstrip(";").split("=")[1])
intro_name = fn
if num_taxa > 1:
tree = bt.loadNewick(input_dir + "/" + treefile,
absoluteTime=False)
#make root line
old_node = tree.root
new_node = bt.node()
new_node.children.append(old_node)
old_node.parent = new_node
old_node.length = 0.000015
new_node.height = 0
new_node.y = old_node.y
tree.root = new_node
tree.Objects.append(new_node)
tips = []
for k in tree.Objects:
if k.branchType == 'leaf':
tips.append(k.name)
if len(tips) < 1000:
df_dict = summarise_node_table(input_dir, treename, taxon_dict)
overall_df_dict[treename] = df_dict
overall_tree_count += 1
make_scaled_tree(tree, treename, input_dir, len(tips),
colour_dict_dict, desired_fields,
tallest_height, taxon_dict, query_dict,
custom_tip_labels, graphic_dict)
else:
too_tall_trees.append(lineage)
continue
return too_tall_trees, overall_tree_count, colour_dict_dict, overall_df_dict
def make_all_of_the_trees(input_dir, taxon_dict, query_id_dict, query_dict, desired_fields, min_uk_taxa=3):
tallest_height = find_tallest_tree(input_dir)
too_tall_trees = []
colour_dict_dict = defaultdict(dict)
overall_tree_count = 0
lst = sort_trees_index(input_dir)
for fn in lst:
lineage = fn
treename = "tree_" + str(fn)
treefile = "tree_" + str(fn) + ".tree"
num_taxa = 0
intro_name = ""
with open(input_dir + "/" + treefile,"r") as f:
for l in f:
l = l.rstrip("\n")
if l.startswith(" Dimensions NTax="):
num_taxa = int(l.rstrip(";").split("=")[1])
intro_name = fn
if num_taxa > 1:
tree = bt.loadNewick(input_dir + "/" + treefile, absoluteTime=False)
old_node = tree.root
new_node = bt.node()
new_node.children.append(old_node)
old_node.parent = new_node
old_node.length=2.0
new_node.height = 0
new_node.y = old_node.y
tree.root = new_node
tree.Objects.append(new_node)
tips = []
for k in tree.Objects:
if k.branchType == 'leaf':
tips.append(k.name)
if len(tips) < 1000:
overall_tree_count += 1
if desired_fields == []:
colour_by = ["adm1"]
else:
colour_by = desired_fields
for trait in colour_by:
colour_dict = find_colour_dict(query_dict, trait)
colour_dict_dict[trait] = colour_dict
make_scaled_tree_without_legend(tree, treename, input_dir, len(tips), colour_dict, trait, tallest_height, lineage, taxon_dict, query_id_dict, query_dict)
else:
too_tall_trees.append(lineage)
continue
return too_tall_trees, overall_tree_count, colour_dict_dict
def make_all_of_the_trees(input_dir,
outdir,
tree_name_stem,
taxon_dict,
query_dict,
colour_fields,
label_fields,
min_uk_taxa=3):
tallest_height = find_tallest_tree(input_dir)
too_tall_trees = []
colour_dict_dict = defaultdict(dict)
overall_df_dict = defaultdict(dict)
overall_tree_count = 0
lst = sort_trees_index(input_dir)
for trait in colour_fields:
colour_dict = find_colour_dict(query_dict, trait)
colour_dict_dict[trait] = colour_dict
for tree_number in lst:
treename = f"tree_{tree_number}"
treefile = f"{tree_name_stem}_{tree_number}.tree"
nodefile = f"{tree_name_stem}_{tree_number}"
num_taxa = 0
tree = bt.loadNewick(input_dir + "/" + treefile, absoluteTime=False)
old_node = tree.root
new_node = bt.node()
new_node.children.append(old_node)
old_node.parent = new_node
old_node.length = 0.000015
new_node.height = 0
new_node.y = old_node.y
tree.root = new_node
tree.Objects.append(new_node)
tips = []
for k in tree.Objects:
if k.branchType == 'leaf':
tips.append(k.name)
if len(tips) < 1000:
df_dict = summarise_node_table(input_dir, nodefile, taxon_dict)
overall_df_dict[treename] = df_dict
overall_tree_count += 1
make_scaled_tree(tree, nodefile, input_dir, outdir, len(tips),
colour_dict_dict, colour_fields, label_fields,
tallest_height, tree_number, taxon_dict,
query_dict)
else:
too_tall_trees.append(tree_number)
continue
return too_tall_trees, overall_tree_count, overall_df_dict, colour_dict_dict
@author: david
"""
from Bio import Phylo
import baltic as bt
import re
# '1-2936','2937-4936', '4937-6870','6871-8473','8474-19706','19707-20428'
in_tree = 'sars-like-CoVs-sub_19707-20428.tre'
out_tree = 'sars-like-CoVs-sub_19707-20428.nexus'
# Basic conversion with Phylo module doesn't work here because we need BEAST-style NEXUS files
#Phylo.convert(in_tree, 'newick', out_tree, 'nexus')
myTree = bt.loadNewick(in_tree, absoluteTime=False)
#myTree.setAbsoluteTime(2019.00) # need to set this to time of last sampled tip
myTree.traverse_tree() ## required to set heights
myTree.treeStats() ## report stats about tree
names = []
for idx, k in enumerate(
myTree.Objects): ## iterate over a flat list of branches
if k.branchType == 'leaf':
curr_name = k.numName
names.append(curr_name)
#print(names)
date_str = '_2020.00'
# Write taxa names
help="Pruned tree including only taxa in list")
args = parser.parse_args()
timetree = args.timetree
tmrcas = args.tmrcas
output = args.output
# path = "/Users/anderson/GLab Dropbox/Anderson Brito/past&future/PhD/works/phylog/species_trees/viral_sppTrees/rhv04_virevol1/trees/ba/run0_host/"
# timetree = path + "host_tree.tree"
# tmrcas = path + 'tmrcas.txt'
# output = path + 'new_host_tree.nexus'
all_traits = ['node_name', 'height_95%_HPD']
# load tree
tree = bt.loadNewick(timetree) #, tip_regex='_([0-9\-]+)$')
# print(tree)
# tmrca dataframe
df = pd.read_csv(tmrcas, encoding='utf-8', sep='\t', dtype='str')
# print(df)
# df['members'] = df['members'].apply(lambda x: ', '.join(sorted(x.split(','))))
# print(df['members'].to_list())
print('Starting tree file processing...')
# transfer supporting value from a newick tree
for k in sorted(tree.Objects, key=lambda q: q.height
): ## iterate over branches from most recent to oldest
if k.branchType == 'node': ## can only sort nodes
terminals = ", ".join(sorted([leaf for leaf in k.leaves]))
