def _get_T(file):
txt = file.read()
lines = txt.split('\n')
lines.pop(0)
lines.pop(len(lines) - 1)
lines = [line.replace('\t', '') for line in lines]
lines = [line.replace(' ', '') for line in lines]
edges = []
edge_labels = {} # key is child node label. val is edge label
for line in lines:
if '->' in line:
i, j = line.split('->')
if '[' in j:
j, edge_label = j.split('[')
edges.append((i, j))
edge_label = '[' + edge_label
edge_labels[j] = edge_label
else:
edges.append((i, j))
edge_labels[j] = ''
parent_child_table = []
for i, j in edges:
if i in list(edge_labels.keys()):
i = i + edge_labels[i]
if j in list(edge_labels.keys()):
j = j + edge_labels[j]
parent_child_table.append((i, j, 1))
return Tree.from_parent_child_table(parent_child_table)
def Get_Clonal_IDs(self):
C_IDs = Tumour_Evolution.keys()
C_Times = []
ID_lenghts = []
Phylogeny = []
C_IDs.remove("P-0:0")
main_branches = []
print "ALL, IDs", C_IDs
for _id in C_IDs:
if (_id.find(",") == -1):
main_branches.append(_id[0:3])
C_IDs.remove(_id)
print "Main, branches ", main_branches
for branch in main_branches:
Phylogeny.append(("P", branch))
print "ph ", Phylogeny
#print "Phylogeny", Phylogeny
print "ids ", C_IDs
for clone in C_IDs:
generations = clone.split(",")
head = generations[0]
last_and_hour = generations[len(generations) - 1].split("-")
last = last_and_hour[0]
time = last_and_hour[1]
generations.remove(generations[0])
generations.remove(generations[len(generations) - 1])
if (not generations):
Phylogeny.append((head, head + "," + last + "-" + time))
else:
print clone
carry = head
for element in generations:
Phylogeny.append((head, head + "," + element))
carry += "," + element
Phylogeny.append((carry, carry + "," + last + "-" + time))
print Phylogeny
p = sorted(list(set(Phylogeny)))
print "p", p
t = Tree.from_parent_child_table(p) #a=np.unique(t).tolist()
ts = TreeStyle()
ts.show_leaf_name = True
#ts.rotation = 90
t.show(tree_style=ts)
def convert(s):
''' Takes in .dot filename as a string, reads through file and converts '''
''' into Newick, which is readable by ete3. '''
nwlist = []
f = open(s)
fl = f.readlines()
for i in fl:
i = i.replace(']', '')
i = i.replace(';', '')
if ('->' in i.split()): #for GT with no branch lengths
nwlist.append((i.split()[0], i.split()[2]))
elif ('--' in i.split()):
nwlist.append((i.split()[0], i.split()[2]))
tree = Tree.from_parent_child_table(nwlist)
return tree
#EOF
from ete3 import Tree, TreeStyle, TextFace
#t = Tree( "((5),0,2,3);", format=1 )
#t = Tree( "(3,5);", format =1 )
t = Tree("(A:1,(B:1,(E:1,D:1):0.5):0.5);")
t = Tree("( ( A,B,(5) ) );", format=1)
t = Tree.from_parent_child_table([("A", "B"), ("A", "C"), ("C", "D"),
("C", "E")])
t = Tree.from_parent_child_table([('P', 'PC1'), ('PC1', 'PC1,2-26:4402'),
('PC1', 'PC1,3'),
('PC1,3', 'PC1,3,5-26:6743'),
('PC1', 'PC1,14'),
('PC1,3,14', 'PC1,3,14,4-27:124'),
('PC1', 'PC1,9'),
('PC1,9', 'PC1,9,7-27:675'),
('PC1', 'PC1,0-26:4132')])
t = Tree.from_parent_child_table([("P", "PC0-23:7392", 100),
("P", "PC1-25:6592", 7.5),
("PC1-25:6592", "PC1,1-26:7008", 6),
("PC1-25:6592", "PC1,2-26:7008", 6)])
ts = TreeStyle()
ts.show_leaf_name = True
#ts.title.add_face(TextFace("Hello ETE", fsize=20), column=0)
t.show(tree_style=ts)
def Surviving_Phylogenetic_Tree(self):
C_IDs = Tumour_Evolution.keys()
C_Times = []
ID_lenghts = []
Phylogeny = []
C_IDs.remove("P-0:0")
main_branches = []
#while loop
regex_str = 'PC[0-9]+'
branch = ',[0-9]+'
entering_flag = True
Phylo_Struct = []
i = 0
while (C_IDs):
if (entering_flag):
entering_flag = False
searchRegex = re.compile(regex_str + '-.*')
matches = [
m.group(0) for l in C_IDs for m in [searchRegex.search(l)]
if m
]
for m in matches:
C_IDs.remove(m)
Phylo_Struct.append(matches)
else:
regex_str = regex_str + branch
searchRegex = re.compile(regex_str + '-.*')
matches = [
m.group(0) for l in C_IDs for m in [searchRegex.search(l)]
if m
]
for m in matches:
C_IDs.remove(m)
Phylo_Struct.append(matches)
#print "PS",Phylo_Struct
# trabnsalte this into a tree
main_branches = Phylo_Struct[0]
branch_ids = []
for clone in main_branches:
branch_ids.append(clone[0:3])
Phylogeny = []
initial_step = Phylo_Struct.pop(0)
initial_step = natsorted(initial_step, key=lambda y: y.lower())
#print "I", initial_step
ID_time = dict()
regex_str = 'PC[0-9]+'
for clone in initial_step:
parent_str = re.search(regex_str, clone)
ID_time[parent_str.group(0)] = int(
clone.split("-", 1)[1].split(":")[0])
Phylogeny.append(
("P", clone, 100 -
int(clone.split("-", 1)[1].split(":")[0]))) ## Year length
#print ID_time
#print "Remaining ", Phylo_Struct
## Generating Phylogenetic Tree
regex_str = 'PC[0-9]+'
branch = ',[0-9]+'
for step in Phylo_Struct:
step = natsorted(step, key=lambda y: y.lower())
for clone in step:
clone_str = re.search(regex_str, clone)
for _parent in initial_step:
parent_str = re.search(regex_str, _parent)
if (clone_str.group(0) == parent_str.group(0)):
main_parent = re.search('PC[0-9]+', clone)
clone_year = int(clone.split("-", 1)[1].split(":")[0])
Phylogeny.append(
(_parent, clone,
abs(ID_time[main_parent.group(0)] -
clone_year))) ## year lengt normalised
regex_str = regex_str + branch
initial_step = step
t = Tree.from_parent_child_table(Phylogeny) #a=np.unique(t).tolist()
ts = TreeStyle()
ts.show_leaf_name = True
#ts.rotation = 90
ts.mode = "c"
ts.arc_start = -180 # 0 degrees = 3 o'clock
ts.arc_span = 180
t.show(tree_style=ts)
