def fusion(self):
get_adjacencies = Extremities_and_adjacencies()
get_chromosomes = Node_evolve.find_chromosomes(self, self.state)
self.linear_chromosomes = get_chromosomes[0]
self.circular_chromosomes = get_chromosomes[1]
chrm_num_1 = random.randint(0, len(self.linear_chromosomes) - 1)
chrm_num_2 = random.randint(0, len(self.linear_chromosomes) - 1)
while chrm_num_1 == chrm_num_2:
chrm_num_2 = random.randint(0, len(self.linear_chromosomes) - 1)
telomeres_chrm1 = [
element for element in self.linear_chromosomes[chrm_num_1]
if type(element) is not tuple
]
telomeres_chrm2 = [
element for element in self.linear_chromosomes[chrm_num_2]
if type(element) is not tuple
]
telo1_num = random.randint(0, len(telomeres_chrm1) - 1)
telo2_num = random.randint(0, len(telomeres_chrm2) - 1)
telo1 = telomeres_chrm1[telo1_num]
telo2 = telomeres_chrm2[telo2_num]
if telo1 < telo2:
new_adj = (telo1, telo2)
else:
new_adj = (telo2, telo1)
# perfrom operation
self.state.remove(telo1)
self.state.remove(telo2)
self.state.append(new_adj)
print('fusion: ', get_adjacencies.adjacencies_to_genome(self.state))
def fission(self):
get_adjacencies = Extremities_and_adjacencies()
get_chromosomes = Node_evolve.find_chromosomes(self, self.state)
self.linear_chromosomes = get_chromosomes[0]
self.circular_chromosomes = get_chromosomes[1]
chrm_num = random.randint(0, len(self.linear_chromosomes) - 1)
adjacencies = [
element for element in self.linear_chromosomes[chrm_num]
if type(element) is tuple
]
while len(adjacencies) == 0:
chrm_num = random.randint(0, len(self.linear_chromosomes) - 1)
adjacencies = [
element for element in self.linear_chromosomes[chrm_num]
if type(element) is tuple
]
adj_num = random.randint(0, len(adjacencies) - 1)
adj = adjacencies[adj_num]
self.state.remove(adj)
self.state.append(adj[0])
self.state.append(adj[1])
print('fission: ', get_adjacencies.adjacencies_to_genome(self.state))
def balanced_translocation(self):
get_adjacencies = Extremities_and_adjacencies()
get_chromosomes = Node_evolve.find_chromosomes(self, self.state)
self.linear_chromosomes = get_chromosomes[0]
self.circular_chromosomes = get_chromosomes[1]
chrm_num_1 = random.randint(0, len(self.linear_chromosomes) - 1)
while len(self.linear_chromosomes[chrm_num_1]) < 3:
chrm_num_1 = random.randint(0, len(self.linear_chromosomes) - 1)
chrm_num_2 = random.randint(0, len(self.linear_chromosomes) - 1)
while chrm_num_1 == chrm_num_2 or len(
self.linear_chromosomes[chrm_num_2]) < 3:
chrm_num_2 = random.randint(0, len(self.linear_chromosomes) - 1)
adjacencies_chrm1 = [
element for element in self.linear_chromosomes[chrm_num_1]
if type(element) is tuple
]
adjacencies_chrm2 = [
element for element in self.linear_chromosomes[chrm_num_2]
if type(element) is tuple
]
adj1_num = random.randint(0, len(adjacencies_chrm1) - 1)
adj2_num = random.randint(0, len(adjacencies_chrm2) - 1)
adj1 = adjacencies_chrm1[adj1_num]
adj2 = adjacencies_chrm2[adj2_num]
if adj1[0] < adj2[1]:
new_adj1 = (adj1[0], adj2[1])
else:
new_adj1 = (adj2[1], adj1[0])
if adj1[1] < adj2[0]:
new_adj2 = (adj1[1], adj2[0])
else:
new_adj2 = (adj2[0], adj1[1])
# perfrom operation
self.state.remove(adj1)
self.state.remove(adj2)
self.state.append(new_adj1)
self.state.append(new_adj2)
print('balanced translocation: ',
get_adjacencies.adjacencies_to_genome(self.state))
def unbalanced_translocation(self):
get_adjacencies = Extremities_and_adjacencies()
get_chromosomes = Node_evolve.find_chromosomes(self, self.state)
self.linear_chromosomes = get_chromosomes[0]
self.circular_chromosomes = get_chromosomes[1]
chrm_num_1 = random.randint(0, len(self.linear_chromosomes) - 1)
# chromosome 1 should not be a single gene chromosome otherwise the operation would amount to a fusion
while len(self.linear_chromosomes[chrm_num_1]) < 3:
chrm_num_1 = random.randint(0, len(self.linear_chromosomes) - 1)
chrm_num_2 = random.randint(0, len(self.linear_chromosomes) - 1)
while chrm_num_1 == chrm_num_2:
chrm_num_2 = random.randint(0, len(self.linear_chromosomes) - 1)
adjacencies_chrm1 = [
element for element in self.linear_chromosomes[chrm_num_1]
if type(element) is tuple
]
telomeres_chrm2 = [
element for element in self.linear_chromosomes[chrm_num_2]
if type(element) is not tuple
]
adj1_num = random.randint(0, len(adjacencies_chrm1) - 1)
adj2_num = random.randint(0, len(telomeres_chrm2) - 1)
adj1 = adjacencies_chrm1[adj1_num]
telo2 = telomeres_chrm2[adj2_num]
if telo2 < adj1[0]:
new_adj1 = (telo2, adj1[0])
else:
new_adj1 = (adj1[0], telo2)
# perfrom operation
self.state.remove(adj1)
self.state.remove(telo2)
self.state.append(new_adj1)
self.state.append(adj1[1])
print('unbaanced translocation: ',
get_adjacencies.adjacencies_to_genome(self.state))
from Class_wrDCJ_Node import Node
from networkx import DiGraph
from Class_extremities_and_adjacencies import Extremities_and_adjacencies
get_genome = Extremities_and_adjacencies()
def build_hash_table(current_node, hash_table, adjacenciesB, weights):
node = current_node
# if the previous operation was a cicularization (i.e. a trp0) do:
if node.join_adjacency != 0:
operations = node.get_reinsertion_operations(adjacenciesB)
for operation in operations:
child_state = node.take_action(operation)[0] # perform operation
check_hash_table = check_hash_key(
child_state, hash_table
) # check whether the intermediate create exists already
# if it is a trp1 type operation
if node.join_adjacency in operation[0]:
operation_type = 'trp1'
operation_weight = 0.5 * weights[1]
#operation_weight = 1 * weights[1]
# else it is a trp2 type operation
else:
from networkx import all_shortest_paths
from Class_wrDCJ_Node import Node
from Class_extremities_and_adjacencies import Extremities_and_adjacencies
from Class_Network_wrDCJ import Network
import GenomeEvolve
number_of_simulations = 1
results = []
genomeB = [[1, 2, 3, 4, 5, 6, 7], [8, 9, 10, 11, 12, 13, 14],
[15, 16, 17, 18, 19, 20, 21], [22, 23, 24, 25, 26, 27, 28],
[29, 30, 31, 32, 33, 34, 35], [36, 37, 38, 39, 40, 41, 42],
[43, 44, 45, 46, 47, 48, 49], [50, 51, 52, 53, 54, 55, 56]]
#genomeB = [[1, 2,3,4, 5, 6, 7],[8,9, 10, 11, 12, 13, 14], [15, 16, 17, 18, 19, 20, 21], [22, 23, 24, 25, 26, 27, 28], [29, 30, 31, 32, 33, 34, 35], [36, 37, 38, 39, 40, 41, 42]]
get_adjacencies = Extremities_and_adjacencies()
adjacencies_genomeB = get_adjacencies.adjacencies_ordered_and_sorted(genomeB)
for i in range(0, number_of_simulations):
genomeB_copy = genomeB.copy()
genomeA = get_adjacencies.adjacencies_to_genome(
GenomeEvolve.evolve_genome(genomeB_copy))
adjacencies_genomeA = get_adjacencies.adjacencies_ordered_and_sorted(
genomeA)
# Create start and target node
start_node = Node(adjacencies_genomeA)
target_node = Node(adjacencies_genomeB)
# Construct entire network
construct_network = Network(start_node, target_node, adjacencies_genomeB)
def build_hash_table(self, current_node):
t0 = time.time()
adjacenciesB = self.adjacenciesB
get_adjacencies = Extremities_and_adjacencies()
#print()
node = current_node
#print()
#print()
#print('current node and state: ')
#print(node)
#print(node.state)
#print('____________________________________________________________________________________')
#if the genome has a circular intermediate (i.e all of its children will be linear)
if node.next_operation != 0:
operation_type = None
operations = []
# if point of cut = previous point of join:
if node.next_operation_weight == 0.5:
operations.append(node.next_operation)
operation_type = 'trp1'
elif node.next_operation_weight == 1.5:
operations = []
if type(node.next_operation) is list:
for operation in node.next_operation:
operations.append(operation)
else:
operations.append(node.next_operation)
operation_type = 'trp2'
else:
print(
'You have got a problem with the .next_operation weights')
for operation in operations:
child_state = node.take_action(operation)[0]
check_hash_table = Network.check_hash_key(self, child_state)
if check_hash_table[0]:
child = check_hash_table[1]
node.children.append(child)
node.children_weights.append(node.next_operation_weight)
node.children_operations.append(
(operation, operation_type))
# print()
# print('Operation: ', operation)
# print('Type: ', operation_type)
# print(get_adjacencies.adjacencies_to_genome(node.state), ' ----> ',
# get_adjacencies.adjacencies_to_genome(child_state))
else:
#remember the child will consist of linear chromosomes only because it is the result of a forced reinsertion
child = Node(child_state)
hash_key = hash(str(child.state))
self.hash_table.update({hash_key: child})
# print('#T: ', self.hash_table)
node.children.append(child)
node.children_weights.append(node.next_operation_weight)
node.children_operations.append(
(operation, operation_type))
# print()
# print('Operation: ', operation)
# print('Type: ', operation_type)
# print(get_adjacencies.adjacencies_to_genome(node.state), ' ----> ',
# get_adjacencies.adjacencies_to_genome(child_state))
Network.build_hash_table(self, child)
#if the genome has no circular intermediates (i.e. some of its children may have circular chromosomes)
else:
operations = node.get_legal_operations(adjacenciesB)
for operation in operations:
operation_result = node.take_action(operation)
child_state = operation_result[0]
op_type = operation_result[1]
check_hash_table = Network.check_hash_key(self, child_state)
if check_hash_table[0]:
child = check_hash_table[1]
node.children.append(child)
# if the operation is a trp0
child.find_chromosomes(child.state)
if len(child.circular_chromosomes) != 0:
node.children_weights.append(0.5)
node.children_operations.append((operation, 'trp0'))
# print()
# print('Operation: ', operation)
# print('Type: ', 'trp0')
# print(get_adjacencies.adjacencies_to_genome(node.state), ' ----> ',
# get_adjacencies.adjacencies_to_genome(child_state))
else:
#node.children_weights.append(1)
#if op_type == 'fis' or op_type == 'fus' or op_type == 'u_trl':
# operation_type = op_type
#else:
# operation_type = node.find_operation_type(operation)
if op_type == 'fis':
operation_type = op_type
op_weight = 2
elif op_type == 'fus':
operation_type = op_type
op_weight = 2
elif op_type == 'u_trl':
operation_type = op_type
op_weight = 1.5
else:
operation_type = node.find_operation_type(
operation)
if operation_type == 'inv':
op_weight = 1
elif operation_type == 'b_trl':
op_weight = 1.5
else:
print(
"There's a problem at the .find_optype node function"
)
node.children_weights.append(op_weight)
node.children_operations.append(
(operation, operation_type))
# print()
# print('Operation: ', operation)
# print('Type: ', operation_type)
# print('weight: ', op_weight)
# print(get_adjacencies.adjacencies_to_genome(node.state), ' ----> ',
# get_adjacencies.adjacencies_to_genome(child_state))
# print(node.children_weights)
else:
child = Node(child_state)
# check whether a circular chromosome has been created
child.find_chromosomes(child.state)
# if a circular chromosome has been created:
if len(child.circular_chromosomes) != 0:
legal_operation = child.get_legal_reinsertion_operation(
operation, self.adjacenciesB)
if legal_operation:
child.next_operation = legal_operation
child.next_operation_weight = 0.5
hash_key = hash(str(child.state))
self.hash_table.update({hash_key: child})
# print('#T: ', self.hash_table)
node.children.append(child)
node.children_operations.append(
(operation, 'trp0'))
node.children_weights.append(0.5)
# print()
# print('Operation: ', operation)
# print('Type: ', 'trp0')
# print(get_adjacencies.adjacencies_to_genome(node.state), ' ----> ',
# get_adjacencies.adjacencies_to_genome(child_state))
Network.build_hash_table(self, child)
else:
child.next_operation = child.get_illegal_decircularization_operation(
self.adjacenciesB)
child.next_operation_weight = 1.5
hash_key = hash(str(child.state))
self.hash_table.update({hash_key: child})
# print('#T: ', self.hash_table)
node.children.append(child)
node.children_operations.append(
(operation, 'trp0'))
node.children_weights.append(0.5)
# print()
# print('Operation: ', operation)
# print('Type: ', 'trp0')
# print(get_adjacencies.adjacencies_to_genome(node.state), ' ----> ',
# get_adjacencies.adjacencies_to_genome(child_state))
Network.build_hash_table(self, child)
# else if no circular chromosome has been created:
else:
hash_key = hash(str(child.state))
self.hash_table.update({hash_key: child})
# print('#T: ', self.hash_table)
node.children.append(child)
'''
if op_type == 'fis' or op_type == 'fus' or op_type == 'u_trl':
operation_type = op_type
else:
operation_type = node.find_operation_type(operation)
node.children_weights.append(1)
node.children_operations.append((operation, operation_type))
print()
print('Operation: ', operation)
print('Type: ', operation_type)
print(get_adjacencies.adjacencies_to_genome(node.stalen(child.circular_chromosomes)te), ' ----> ',
get_adjacencies.adjacencies_to_genome(child_state))
'''
if op_type == 'fis':
operation_type = op_type
op_weight = 2
elif op_type == 'fus':
operation_type = op_type
op_weight = 2
elif op_type == 'u_trl':
operation_type = op_type
op_weight = 1.5
else:
operation_type = node.find_operation_type(
operation)
if operation_type == 'inv':
op_weight = 1
elif operation_type == 'b_trl':
op_weight = 1.5
else:
print(
"There's a problem at the .find_optype node function"
)
node.children_weights.append(op_weight)
node.children_operations.append(
(operation, operation_type))
Network.build_hash_table(self, child)
def evolve(self,
num_inv=0,
num_fus=0,
num_fis=0,
num_b_trl=0,
num_u_trl=0,
num_trp1=0,
num_trp2=0):
get_adjacencies = Extremities_and_adjacencies()
total = num_b_trl + num_fis + num_fus + num_inv + num_trp1 + num_trp2 + num_u_trl
inv_n = num_inv
fus_n = num_fus
fis_n = num_fis
b_trl_n = num_b_trl
u_trl_n = num_u_trl
trp1_n = num_trp1
trp2_n = num_trp2
operations = ['inv', 'trp1', 'trp2', 'b_trl', 'u_trl', 'fis', 'fus']
while total > 0:
operation = random.choice(operations)
if operation == 'inv':
if inv_n > 0:
# print('current state = ', get_adjacencies.adjacencies_to_genome(self.state))
self.inversion()
inv_n -= 1
total -= 1
else:
pass
elif operation == 'trp1':
if trp1_n > 0:
# print('current state = ', get_adjacencies.adjacencies_to_genome(self.state))
self.transposition1()
trp1_n -= 1
total -= 1
else:
pass
elif operation == 'trp2':
if trp2_n > 0:
# print('current state = ', get_adjacencies.adjacencies_to_genome(self.state))
self.transposition2()
trp2_n -= 1
total -= 1
else:
pass
elif operation == 'b_trl':
if b_trl_n > 0:
# print('current state = ', get_adjacencies.adjacencies_to_genome(self.state))
self.balanced_translocation()
b_trl_n -= 1
total -= 1
else:
pass
elif operation == 'u_trl':
if u_trl_n > 0:
# print('current state = ', get_adjacencies.adjacencies_to_genome(self.state))
self.unbalanced_translocation()
u_trl_n -= 1
total -= 1
else:
pass
elif operation == 'fis':
if fis_n > 0:
# print('current state = ', get_adjacencies.adjacencies_to_genome(self.state))
self.fission()
fis_n -= 1
total -= 1
else:
pass
elif operation == 'fus':
if fus_n > 0:
# print('current state = ', get_adjacencies.adjacencies_to_genome(self.state))
self.fusion()
fus_n -= 1
total -= 1
else:
pass
def inversion(self):
get_adjacencies = Extremities_and_adjacencies()
get_chromosomes = Node_evolve.find_chromosomes(self, self.state)
self.linear_chromosomes = get_chromosomes[0]
self.circular_chromosomes = get_chromosomes[1]
chrm_num = random.randint(0, len(self.linear_chromosomes) - 1)
# if it is a single gene or two gene chromosome, choose another one
while len(self.linear_chromosomes[chrm_num]) < 4:
chrm_num = random.randint(0, len(self.linear_chromosomes) - 1)
adjacencies = [
element for element in self.linear_chromosomes[chrm_num]
if type(element) is tuple
]
adj1_num = random.randint(0, len(adjacencies) - 1)
adj2_num = random.randint(0, len(adjacencies) - 1)
while adj1_num == adj2_num:
adj2_num = random.randint(0, len(adjacencies) - 1)
adj1 = adjacencies[adj1_num]
adj2 = adjacencies[adj2_num]
if adj1[0] < adj2[0]:
new_adj1 = (adj1[0], adj2[0])
else:
new_adj1 = (adj2[0], adj1[0])
if adj1[1] < adj2[1]:
new_adj2 = (adj1[1], adj2[1])
else:
new_adj2 = (adj2[1], adj1[1])
#perfrom operation
self.state.remove(adj1)
self.state.remove(adj2)
self.state.append(new_adj1)
self.state.append(new_adj2)
get_chromosomes = Node_evolve.find_chromosomes(self, self.state)
self.linear_chromosomes = get_chromosomes[0]
self.circular_chromosomes = get_chromosomes[1]
while len(self.circular_chromosomes) > 0:
self.state.append(adj1)
self.state.append(adj2)
self.state.remove(new_adj1)
self.state.remove(new_adj2)
get_chromosomes = Node_evolve.find_chromosomes(self, self.state)
self.linear_chromosomes = get_chromosomes[0]
self.circular_chromosomes = get_chromosomes[1]
chrm_num = random.randint(0, len(self.linear_chromosomes) - 1)
while len(self.linear_chromosomes[chrm_num]) < 4:
chrm_num = random.randint(0, len(self.linear_chromosomes) - 1)
adjacencies = [
element for element in self.linear_chromosomes[chrm_num]
if type(element) is tuple
]
adj1_num = random.randint(0, len(adjacencies) - 1)
adj2_num = random.randint(0, len(adjacencies) - 1)
while adj1_num == adj2_num:
adj2_num = random.randint(0, len(adjacencies) - 1)
adj1 = adjacencies[adj1_num]
adj2 = adjacencies[adj2_num]
if adj1[0] < adj2[0]:
new_adj1 = (adj1[0], adj2[0])
else:
new_adj1 = (adj2[0], adj1[0])
if adj1[1] < adj2[1]:
new_adj2 = (adj1[1], adj2[1])
else:
new_adj2 = (adj2[1], adj1[1])
# perfrom operation
self.state.remove(adj1)
self.state.remove(adj2)
self.state.append(new_adj1)
self.state.append(new_adj2)
get_chromosomes = Node_evolve.find_chromosomes(self, self.state)
self.linear_chromosomes = get_chromosomes[0]
self.circular_chromosomes = get_chromosomes[1]
print('inversion: ',
get_adjacencies.adjacencies_to_genome(self.state))
def transposition(self, linear_chromosomes):
get_adjacencies_and_chromosomes = Extremities_and_adjacencies()
# print('CHROMOSOMES Before transposition 1: ', get_adjacencies_and_chromosomes.find_chromosomes(self.state))
get_adjacencies_and_chromosomes = Extremities_and_adjacencies()
chromosome_number = random.randint(0, len(linear_chromosomes) - 1)
chromosome = linear_chromosomes[chromosome_number]
chromosome_length = len(chromosome)
to_exclude = []
while chromosome_length < 4:
to_exclude.append(linear_chromosomes.index(chromosome))
# print('Transposition a')
# print('THE LIST: ', [i for i in range(0, len(linear_chromosomes)) if i not in to_exclude])
# print(linear_chromosomes)
chromosome_number = random.choice([
i for i in range(0, len(linear_chromosomes))
if i not in to_exclude
])
chromosome = linear_chromosomes[chromosome_number]
chromosome_length = len(chromosome)
try:
adjacencies = [
element for element in chromosome if type(element) is tuple
]
except:
print(
'There were not any chromosomes with more than 2 sequence blocks'
)
adj1_num = random.randint(0, len(adjacencies) - 1)
adj2_num = random.randint(0, len(adjacencies) - 1)
while adj1_num == adj2_num:
adj2_num = random.randint(0, len(adjacencies) - 1)
adj1 = adjacencies[adj1_num]
adj2 = adjacencies[adj2_num]
# order extremities in the adjacency tuple
if adj1[0] < adj2[0]:
new_adj1 = (adj1[0], adj2[0])
else:
new_adj1 = (adj2[0], adj1[0])
if adj1[1] < adj2[1]:
new_adj2 = (adj1[1], adj2[1])
else:
new_adj2 = (adj2[1], adj1[1])
# perform_operation
self.state.remove(adj1)
self.state.remove(adj2)
self.state.append(new_adj1)
self.state.append(new_adj2)
chromosomes = get_adjacencies_and_chromosomes.find_chromosomes(
self.state)
if len(chromosomes[1]) == 0:
# reverse operation
self.state.append(adj1)
self.state.append(adj2)
self.state.remove(new_adj1)
self.state.remove(new_adj2)
# order extremities in the adjacency tuple
if adj1[0] < adj2[1]:
new_adj1 = (adj1[0], adj2[1])
else:
new_adj1 = (adj2[1], adj1[0])
if adj1[1] < adj2[0]:
new_adj2 = (adj1[1], adj2[0])
else:
new_adj2 = (adj2[0], adj1[1])
# perform_operation
self.state.remove(adj1)
self.state.remove(adj2)
self.state.append(new_adj1)
self.state.append(new_adj2)
# get operation notation
if adj1[0] < adj2[0]:
operationA = (adj1, adj2)
else:
operationA = (adj2, adj1)
if new_adj1[0] < new_adj2[0]:
operationB = (new_adj1, new_adj2)
else:
operationB = (new_adj2, new_adj1)
chromosomes = get_adjacencies_and_chromosomes.find_chromosomes(
self.state)
if new_adj1 in chromosomes[1][0]:
circular_adjacency = new_adj1
exsision_adjacency = new_adj2
else:
circular_adjacency = new_adj2
exsision_adjacency = new_adj1
operation1 = ((operationA, operationB), 'trp0')
intermediate1 = self.state[:]
###############
# Reinsertion DCJ
adj1 = circular_adjacency
adj2 = exsision_adjacency
counter = 0
while adj2 == exsision_adjacency:
if counter > 2:
adj2 = chromosomes[0][0][0]
else:
chromosome_number = random.randint(0, len(chromosomes[0]) - 1)
chromosome = chromosomes[0][chromosome_number]
chromosome_length = len(chromosome)
to_exclude = []
while chromosome_length < 3:
to_exclude.append(linear_chromosomes.index(chromosome))
# print('Transposition b')
## print('THE LIST: ', [i for i in range(0, len(linear_chromosomes)) if i not in to_exclude])
# print(linear_chromosomes)
chromosome_number = random.choice([
i for i in range(0, len(linear_chromosomes))
if i not in to_exclude
])
chromosome = chromosomes[0][chromosome_number]
chromosome_length = len(chromosome)
adjacencies = [
element for element in chromosome if type(element) is tuple
]
adj2_num = random.randint(0, len(adjacencies) - 1)
adj2 = adjacencies[adj2_num]
counter += 1
# order extremities in adjacency tuple
if type(adj2) is not tuple:
if adj1[0] < adj2:
new_adj1 = (adj1[0], adj2)
else:
new_adj1 = (adj2, adj1[0])
new_adj2 = adj1[1]
# perfrom operation
self.state.remove(adj1)
self.state.remove(adj2)
self.state.append(new_adj1)
self.state.append(new_adj2)
if adj1[0] < adj2:
operationA = (adj1, adj2)
else:
operationA = (adj2, adj1)
if new_adj1[0] < new_adj2:
operationB = (new_adj1, new_adj2)
else:
operationB = (new_adj2, new_adj1)
operation2 = ((operationA, operationB), 'trp1')
intermediate2 = self.state[:]
else:
try:
if adj1[0] < adj2[0]:
new_adj1 = (adj1[0], adj2[0])
else:
new_adj1 = (adj2[0], adj1[0])
except TypeError:
print("transcroption problem in line 293")
print('adj1: ', adj1)
print('adj2: ', adj2)
if adj1[1] < adj2[1]:
new_adj2 = (adj1[1], adj2[1])
else:
new_adj2 = (adj2[1], adj1[1])
# perfrom operation
self.state.remove(adj1)
self.state.remove(adj2)
self.state.append(new_adj1)
self.state.append(new_adj2)
if adj1[0] < adj2[0]:
operationA = (adj1, adj2)
else:
operationA = (adj2, adj1)
if new_adj1[0] < new_adj2[0]:
operationB = (new_adj1, new_adj2)
else:
operationB = (new_adj2, new_adj1)
operation2 = ((operationA, operationB), 'trp1')
intermediate2 = self.state[:]
return [(operation1, intermediate1), (operation2, intermediate2)]
def __init__(self, target_genome):
get_adjacencies_and_chromosomes = Extremities_and_adjacencies()
self.state = get_adjacencies_and_chromosomes.create_adjacency_list(
target_genome)
def inversion(self, linear_chromosomes):
get_adjacencies_and_chromosomes = Extremities_and_adjacencies()
chromosome_number = random.randint(0, len(linear_chromosomes) - 1)
chromosome = linear_chromosomes[chromosome_number]
#print('the chromosome: ', chromosome)
chromosome_length = len(chromosome)
#print('chromosome length: ', chromosome_length)
to_exclude = []
# print('CHROMOSOMES Before Inversion: ', get_adjacencies_and_chromosomes.find_chromosomes(self.state))
while chromosome_length < 3:
to_exclude.append(linear_chromosomes.index(chromosome))
# print('Inversion')
# print('THE LIST: ', [i for i in range(0, len(linear_chromosomes)) if i not in to_exclude])
# print(linear_chromosomes)
chromosome_number = random.choice([
i for i in range(0, len(linear_chromosomes))
if i not in to_exclude
])
chromosome = linear_chromosomes[chromosome_number]
chromosome_length = len(chromosome)
if chromosome_length == 3:
#print('is executing')
adjacencies = [
element for element in chromosome if type(element) is tuple
]
telomeres = [
element for element in chromosome if type(element) is not tuple
]
adj = adjacencies[0]
telo = random.choice(telomeres)
if int(telo) == int(adj[0]):
if telo < adj[1]:
new_adj = (telo, adj[1])
else:
new_adj = (adj[1], telo)
new_telo = adj[0]
else:
if telo < adj[0]:
new_adj = (telo, adj[0])
else:
new_adj = (adj[0], telo)
new_telo = adj[1]
operationA = (adj, telo)
operationB = (new_adj, new_telo)
self.state.remove(adj)
self.state.remove(telo)
self.state.append(new_adj)
self.state.append(new_telo)
operation = (operationA, operationB, 'inv')
intermediate = self.state
# print('operation: ', operation)
## print('done')
else:
try:
adjacencies = [
element for element in chromosome if type(element) is tuple
]
except:
print(
'There were not any chromosomes with more than 2 sequence blocks'
)
adj1_num = random.randint(0, len(adjacencies) - 1)
adj2_num = random.randint(0, len(adjacencies) - 1)
while adj1_num == adj2_num:
adj2_num = random.randint(0, len(adjacencies) - 1)
adj1 = adjacencies[adj1_num]
adj2 = adjacencies[adj2_num]
#order extremities in the adjacency tuple
if adj1[0] < adj2[0]:
new_adj1 = (adj1[0], adj2[0])
else:
new_adj1 = (adj2[0], adj1[0])
if adj1[1] < adj2[1]:
new_adj2 = (adj1[1], adj2[1])
else:
new_adj2 = (adj2[1], adj1[1])
#perform_operation
self.state.remove(adj1)
self.state.remove(adj2)
self.state.append(new_adj1)
self.state.append(new_adj2)
chromosomes = get_adjacencies_and_chromosomes.find_chromosomes(
self.state)
if len(chromosomes[1]) != 0:
#reverse operation
self.state.append(adj1)
self.state.append(adj2)
self.state.remove(new_adj1)
self.state.remove(new_adj2)
# order extremities in the adjacency tuple
if adj1[0] < adj2[1]:
new_adj1 = (adj1[0], adj2[1])
else:
new_adj1 = (adj2[1], adj1[0])
if adj1[1] < adj2[0]:
new_adj2 = (adj1[1], adj2[0])
else:
new_adj2 = (adj2[0], adj1[1])
# perform_operation
self.state.remove(adj1)
self.state.remove(adj2)
self.state.append(new_adj1)
self.state.append(new_adj2)
# get operation notation
if adj1[0] < adj2[0]:
operationA = (adj1, adj2)
else:
operationA = (adj2, adj1)
if new_adj1[0] < new_adj2[0]:
operationB = (new_adj1, new_adj2)
else:
operationB = (new_adj2, new_adj1)
operation = ((operationA, operationB), 'inv')
intermediate = self.state[:]
# print('CHROMOSOMES After: ', get_adjacencies_and_chromosomes.find_chromosomes(self.state))
# print('operation: ',operation)
return (operation, intermediate)
def evolve_with_random_rearrangements(self, number_of_rearrangements):
rearrangements = ['inv', 'trp', 'b_trl', 'u_trl', 'fus', 'fis']
get_adjacencies_and_chromosomes = Extremities_and_adjacencies()
#genome_adjacencies = get_adjacencies_and_chromosomes.create_adjacency_list(target_genome)
rearrangement_series = []
while number_of_rearrangements > 0:
#print('NUMBER OF REARRANGEMENTS: ', number_of_rearrangements)
chromosomes = get_adjacencies_and_chromosomes.find_chromosomes(
self.state)
linear_chromosomes = chromosomes[0]
if len(chromosomes[0]) < 2: #if it is a single chromosome genome
if len(chromosomes[0][0]) < 5:
operation = 'inv'
else:
chr_lens = [len(x) for x in linear_chromosomes]
if max(chr_lens) < 4:
operation = 'inv'
else:
operation = random.choice(['inv', 'trp'])
# elif len(chromosomes[0]) == 2 and (len(chromosomes[0][0])==2 or len(chromosomes[0][1])==2): # if the genome consists of two chromosomes but one chromosomes is a single gene chromosome
# operation = random.choice(['inv', 'trp', 'u_trl', 'fis', 'fus'])
else:
chr_lens = [len(x) for x in linear_chromosomes]
chrm_longer_than_2_SB = [x for x in chr_lens if x > 3]
chrm_longer_than_1_SB = [x for x in chr_lens if x > 2]
if len(chrm_longer_than_2_SB) < 2: #cannot be trp
if len(chrm_longer_than_1_SB) < 2: #cannot be b_trl:
if len(chrm_longer_than_1_SB
) < 1: #cannot be u_trl, fis, inv
operation = 'fus'
else:
operation = random.choice(
['inv', 'u_trl', 'fus', 'fis'])
else:
operation = random.choice(
['inv', 'b_trl', 'u_trl', 'fus', 'fis'])
else:
operation = random.choice(rearrangements)
# if max(chr_lens) < 4: #cannot be trp
# number_of_chromosomes_with_more_than_one_sequence_block = 0
# for element in chr_lens:
# if element > 2:
# number_of_chromosomes_with_more_than_one_sequence_block+=1
# if number_of_chromosomes_with_more_than_one_sequence_block >= 2:
# operation = random.choice(['inv', 'b_trl' ,'u_trl', 'fis', 'fus'])
# else:
# operation = random.choice(['inv', 'u_trl', 'fis', 'fus'])
# else:
# number_of_chromosomes_with_more_than_one_sequence_block = 0
# for element in chr_lens:
# if element > 2:
# number_of_chromosomes_with_more_than_one_sequence_block += 1
# if number_of_chromosomes_with_more_than_one_sequence_block >= 2:
# operation = random.choice(['inv', 'trp','b_trl', 'u_trl', 'fis', 'fus'])
# else:
# operation = random.choice(['inv', 'trp','u_trl', 'fis', 'fus'])
#
# number_of_chromosomes_with_more_than_one_sequence_block = 0
#
# for element in chromosomes[0]:
# if len(element) > 2:
# number_of_chromosomes_with_more_than_one_sequence_block+=1
# if number_of_chromosomes_with_more_than_one_sequence_block >= 2:
# operation = random.choice(rearrangements)
# else:
# operation = random.choice(['inv', 'trp', 'u_trl', 'fis', 'fus'])
if operation == 'inv':
execution = self.inversion(linear_chromosomes)
#print('REACHES THIS POINT')
rearrangement_series.append(execution)
number_of_rearrangements -= 1
#print('NUMBER OF REARRANGEMENTS after: ', number_of_rearrangements)
elif operation == 'trp':
execution = self.transposition(linear_chromosomes)
for element in execution:
rearrangement_series.append(element)
number_of_rearrangements -= 1
elif operation == 'b_trl':
execution = self.balanced_translocation(linear_chromosomes)
rearrangement_series.append(execution)
number_of_rearrangements -= 1
elif operation == 'u_trl':
execution = self.unbalanced_translocation(linear_chromosomes)
rearrangement_series.append(execution)
number_of_rearrangements -= 1
elif operation == 'fus':
execution = self.fusion(linear_chromosomes)
rearrangement_series.append(execution)
number_of_rearrangements -= 1
elif operation == 'fis':
execution = self.fission(linear_chromosomes)
rearrangement_series.append(execution)
number_of_rearrangements -= 1
return rearrangement_series
def run(args):
genomeA_file = args.source_genome
genomeB_file = args.target_genome
weight_ratios_file = args.ratios
stdoutOrigin = sys.stdout
sys.stdout = open(args.output_file, 'w')
#outfile = open(args.output_file, 'w')
with open(genomeA_file) as csv:
line = [element.strip('\n').split(',') for element in csv]
genomeA = []
for element in line:
element = list(map(int, element))
genomeA.append(element)
with open(genomeB_file) as csv:
line = [element.strip('\n').split(',') for element in csv]
genomeB = []
for element in line:
element = list(map(int, element))
genomeB.append(element)
with open(weight_ratios_file) as csv:
line = [element.strip('\n').split(',') for element in csv]
weight_ratios = []
for element in line:
element = list(map(int, element))
weight_ratios.append(element)
get_adjacencies = Extremities_and_adjacencies()
adjacencies_genomeA = get_adjacencies.adjacencies_ordered_and_sorted(genomeA)
adjacencies_genomeB = get_adjacencies.adjacencies_ordered_and_sorted(genomeB)
#Create start and target node
start_node = Node(adjacencies_genomeA)
target_node = Node(adjacencies_genomeB)
hash_table = {}
hash_key_start = hash(str(start_node.state))
hash_key_target = hash(str(target_node.state))
hash_table.update({hash_key_start:start_node})
hash_table.update({hash_key_target:target_node})
#finding rearrangement weights
max_number = max(weight_ratios[0])
weights = []
for element in weight_ratios[0]:
if element == 0:
weights.append(max_number^2)
else:
weights.append(max_number/element)
New_Network_wrDCJ.build_hash_table(start_node, hash_table, adjacencies_genomeB, weights)
network = New_Network_wrDCJ.build_network(hash_table)
shortest_paths = (list(all_shortest_paths(network, start_node, target_node, weight='weight')))
j = 1
tot_b_trl = 0
tot_u_trl = 0
tot_inv = 0
tot_trp1 = 0
tot_trp2 = 0
tot_fus = 0
tot_fis = 0
Paths_state = []
Paths_state_weight = []
# print(shortest_paths[0][4].children_weights[2])
for path in shortest_paths:
path_state = []
path_state_weight = []
i = 0
while i < len(path):
current = path[i]
if i == 0:
operation_type = 'none, this is the source genome'
operation_weight = 'N/A'
operation = 'N/A'
else:
x = path[i - 1].children.index(current)
operation_type = path[i - 1].children_operations[x][1]
operation_weight = path[i - 1].children_weights[x]
operation = path[i - 1].children_operations[x][0]
adjacencies = current.state
genome = get_adjacencies.adjacencies_to_genome(adjacencies)
path_state_weight.append((genome, ((operation_type, operation), operation_weight)))
path_state.append((genome, (operation_type, operation)))
i += 1
Paths_state.append((path_state))
Paths_state_weight.append(path_state_weight)
for path in shortest_paths:
i = 0
b_trl = 0
u_trl = 0
inv = 0
trp1 = 0
trp2 = 0
fus = 0
fis = 0
while i < len(path):
current = path[i]
if i == 0:
pass
else:
x = path[i - 1].children.index(current)
operation_type = path[i - 1].children_operations[x][1]
if operation_type == 'b_trl':
b_trl += 1
elif operation_type == 'u_trl':
u_trl += 1
elif operation_type == 'inv':
inv += 1
elif operation_type == 'trp1':
trp1 += 1
elif operation_type == 'trp2':
trp2 += 1
elif operation_type == 'fus':
fus += 1
elif operation_type == 'fis':
fis += 1
i += 1
tot_b_trl += b_trl
tot_u_trl += u_trl
tot_inv += inv
tot_trp1 += trp1
tot_trp2 += trp2
tot_fus += fus
tot_fis += fis
j += 1
print('############################################################################################################')
print()
print('Source Genome: ', genomeA)
print('Target Genome: ', genomeB)
print()
print('Number of most parsimonious solutions: ', len(shortest_paths))
print()
print('Average number of each operation per solution:')
print('Inversions: ', int(tot_inv/len(shortest_paths)), ' Transpositions type 1: ', int(tot_trp1/len(shortest_paths)), ' Transpositions type 2: ', int(tot_trp2/len(shortest_paths)), ' Balanced translocations: ', int(tot_b_trl/len(shortest_paths)), ' Unbalanced translocations: ', int(tot_u_trl/len(shortest_paths)),
' Fusions: ', int(tot_fus/len(shortest_paths)),
' Fissions: ', int(tot_fis/len(shortest_paths)))
print()
print()
print('Solutions: ')
print()
path_counter = 1
for path in Paths_state:
print('Solution number ', path_counter)
for genome in path:
print(genome)
path_counter+=1
print()
print()
print('############################################################################################################')
###############################
# JUST FOR TESTING
solution = [([[1, 2, 3, 4, 15], [-8, -7, 6, -5, -14, -13, -12], [9, 11],
[-20, -19, -18, -17, -16, -32, 10, -31, -30, -29, -28, -27], [21, 22, 23, 24, 25, 26], [-33],
[34, 35, 36, 37, 38, 39, 40]], ('none, this is the source genome', 'N/A')), (
[[1, 2, 3, 4, 15], [-8, -7, -6, -5, -14, -13, -12], [9, 11],
[-20, -19, -18, -17, -16, -32, 10, -31, -30, -29, -28, -27], [21, 22, 23, 24, 25, 26], [-33],
[34, 35, 36, 37, 38, 39, 40]], ('inv', (((5.5, 6.5), (6, 7)), ((5.5, 6), (6.5, 7))))), (
[[1, 2, 3, 4, 15], [-8, -7, -6, -5, -14, -13, -12], [9, 11], [16, 17, 18, 19, 20],
[21, 22, 23, 24, 25, 26], [27, 28, 29, 30, 31, -10, 32, 33], [34, 35, 36, 37, 38, 39, 40]],
('u_trl', (((16, 32.5), 33), ((32.5, 33), 16)))), (
[[1, 2, 3, 4, 5, 6, 7, 8], [9, 11], [12, 13, 14, 15], [16, 17, 18, 19, 20], [21, 22, 23, 24, 25, 26],
[27, 28, 29, 30, 31, -10, 32, 33], [34, 35, 36, 37, 38, 39, 40]],
('b_trl', (((4.5, 15), (5, 14.5)), ((4.5, 5), (14.5, 15))))), (
[[1, 2, 3, 4, 5, 6, 7, 8], [9, 11], [12, 13, 14, 15], [16, 17, 18, 19, 20], [21, 22, 23, 24, 25, 26],
[27, 28, 29, 30, 31, 32, 33], [34, 35, 36, 37, 38, 39, 40], ['o', 10]],
('trp0', (((10, 32), (10.5, 31.5)), ((10, 10.5), (31.5, 32))))), (
[[1, 2, 3, 4, 5, 6, 7, 8], [9, 10, 11], [12, 13, 14, 15], [16, 17, 18, 19, 20],
[21, 22, 23, 24, 25, 26], [27, 28, 29, 30, 31, 32, 33], [34, 35, 36, 37, 38, 39, 40]],
('trp1', (((9.5, 11), (10, 10.5)), ((9.5, 10), (10.5, 11)))))]
paths_operations = []
for element in Paths_state:
path_operations = [y for (x, y) in element]
paths_operations.append(path_operations)
solution_operations = [d for (c, d) in solution]
path_types = []
sol_types = [a for a, b in solution_operations]
for element in paths_operations:
types = [c for c, d in element]
path_types.append(types)
indexes = []
counter = 0
for element in path_types:
if element == sol_types:
indexer = path_types.index(element)
counter += 1
indexes.append(indexer)
# for element in indexes:
# for x in Paths_state[element]:
# print(x)
# print()
# print('*****')
# for element in solution:
# print(element)
print('sol len: ', len(solution))
print('shortest path len: ',len(shortest_paths[0]))
print('counter', counter)
print('And the answer is... ', solution_operations in paths_operations)
print('Source genome: ',genomeA)
print('Target genome: ', genomeB)
print()
print('Solution: ', solution)
##########################################################################################################
sys.stdout.close()
sys.stdout=stdoutOrigin
from Class_extremities_and_adjacencies import Extremities_and_adjacencies
from Class_Evolve import Node_evolve
get_adjacencies = Extremities_and_adjacencies()
'''
genomeA = [[1,2,4,3]]
genomeB = [[1, 2,3,4, 5,6,7],[8,9, 10, 11, 12], [13, 14, 15]]
adjacencies_genomeA = get_adjacencies.adjacencies_ordered_and_sorted(genomeA)
adjacencies_genomeB = get_adjacencies.adjacencies_ordered_and_sorted(genomeB)
print('Adjacencies of the genomes: ')
print('Genome A: ', adjacencies_genomeA)
print('Genome B: ', adjacencies_genomeB)
print('____________________________________')
print()
print()
genome = Node_evolve(state=adjacencies_genomeB)
print(get_adjacencies.adjacencies_to_genome(genome.state))
print()
genome.evolve(num_inv=1, num_b_trl=1, num_fis=1, num_fus=1, num_trp1=1, num_trp2=0, num_u_trl=1)
print()
print(get_adjacencies.adjacencies_to_genome(genome.state))
'''
def trp2(self):
get_adjacencies = Extremities_and_adjacencies()
# exsision and circularization
get_chromosomes = Node_evolve.find_chromosomes(self, self.state)
self.linear_chromosomes = get_chromosomes[0]
self.circular_chromosomes = get_chromosomes[1]
chrm_num = random.randint(0, len(self.linear_chromosomes) - 1)
# if it is a single gene or two gene chromosome, choose another one
while len(self.linear_chromosomes[chrm_num]) < 4:
chrm_num = random.randint(0, len(self.linear_chromosomes) - 1)
# if there is only one adjacency
adjacencies = [
element for element in self.linear_chromosomes[chrm_num]
if type(element) is tuple
]
adj1_num = random.randint(0, len(adjacencies) - 1)
adj2_num = random.randint(0, len(adjacencies) - 1)
while adj1_num == adj2_num:
adj2_num = random.randint(0, len(adjacencies) - 1)
adj1 = adjacencies[adj1_num]
adj2 = adjacencies[adj2_num]
if adj1[0] < adj2[1]:
new_adj1 = (adj1[0], adj2[1])
else:
new_adj1 = (adj2[1], adj1[0])
if adj1[1] < adj2[0]:
new_adj2 = (adj1[1], adj2[0])
else:
new_adj2 = (adj2[0], adj1[1])
# perfrom operation
self.state.remove(adj1)
self.state.remove(adj2)
self.state.append(new_adj1)
self.state.append(new_adj2)
get_chromosomes = Node_evolve.find_chromosomes(self, self.state)
self.linear_chromosomes = get_chromosomes[0]
self.circular_chromosomes = get_chromosomes[1]
if len(self.circular_chromosomes) == 0:
self.state.append(adj1)
self.state.append(adj2)
self.state.remove(new_adj1)
self.state.remove(new_adj2)
if adj1[0] < adj2[0]:
new_adj1 = (adj1[0], adj2[0])
else:
new_adj1 = (adj2[0], adj1[0])
if adj1[1] < adj2[1]:
new_adj2 = (adj1[1], adj2[1])
else:
new_adj2 = (adj2[1], adj1[1])
self.state.remove(adj1)
self.state.remove(adj2)
self.state.append(new_adj1)
self.state.append(new_adj2)
get_chromosomes = Node_evolve.find_chromosomes(self, self.state)
self.linear_chromosomes = get_chromosomes[0]
self.circular_chromosomes = get_chromosomes[1]
if new_adj1 in self.circular_chromosomes[0]:
join = new_adj1
excision = new_adj2
else:
join = new_adj2
excision = new_adj1
print('trp0: ', get_adjacencies.adjacencies_to_genome(self.state))
# decircularization and reinsertion
adj1 = join
chrm_num = random.randint(0, len(self.linear_chromosomes) - 1)
# if it is a single gene chromosome, choose another chromosome
while len(self.linear_chromosomes[chrm_num]) == 2:
chrm_num = random.randint(0, len(self.linear_chromosomes) - 1)
if len(self.linear_chromosomes[chrm_num]) == 3:
adjacencies = [
element for element in self.linear_chromosomes[chrm_num]
if type(element) is tuple
]
adj2 = adjacencies[0]
else:
adjacencies = [
element for element in self.linear_chromosomes[chrm_num]
if type(element) is tuple
]
adj2_num = random.randint(0, len(adjacencies) - 1)
adj2 = adjacencies[adj2_num]
while adj2 == excision:
chrm_num = random.randint(0, len(self.linear_chromosomes) - 1)
# if it is a single gene chromosome, choose another chromosome
while len(self.linear_chromosomes[chrm_num]) == 2:
chrm_num = random.randint(0, len(self.linear_chromosomes) - 1)
if len(self.linear_chromosomes[chrm_num]) == 3:
adjacencies = [
element for element in self.linear_chromosomes[chrm_num]
if type(element) is tuple
]
adj2 = adjacencies[0]
else:
adjacencies = [
element for element in self.linear_chromosomes[chrm_num]
if type(element) is tuple
]
adj2_num = random.randint(0, len(adjacencies) - 1)
adj2 = adjacencies[adj2_num]
if adj1[0] < adj2[1]:
new_adj1 = (adj1[0], adj2[1])
else:
new_adj1 = (adj2[1], adj1[0])
if adj1[1] < adj2[0]:
new_adj2 = (adj1[1], adj2[0])
else:
new_adj2 = (adj2[0], adj1[1])
# perfrom operation
self.state.remove(adj1)
self.state.remove(adj2)
self.state.append(new_adj1)
self.state.append(new_adj2)
print('trp1: ', get_adjacencies.adjacencies_to_genome(self.state))
from networkx import all_shortest_paths
from Class_wrDCJ_Node import Node
from Class_extremities_and_adjacencies import Extremities_and_adjacencies
import New_Network_wrDCJ
import GenomeEvolve
import time
t0 = time.time()
number_of_simulations = 10000
results = []
genomeB = [[1,2,3,4,5,6,7,8,9,10], [11,12,13,14,15,16,17,18,19,20], [21, 22,23,24,25,26,27,28,29,30], [31,32,33,34,35,36,37,38,39,40], [41,42,43,44,45,46,47,48,49,50]]
weight_ratios = [[1,1,1,1,1,1]]
get_adjacencies = Extremities_and_adjacencies()
adjacencies_genomeB = get_adjacencies.adjacencies_ordered_and_sorted(genomeB)
number_of_solutions_found = 0
for i in range(0, number_of_simulations):
print('simultation: ', i)
genomeB_copy = genomeB[:]
evolution_simulation = GenomeEvolve.get_evolved_genome_and_solution(genomeB_copy)
genomeA = evolution_simulation[1]
sorting_scenario = evolution_simulation[2]
adjacencies_genomeA = get_adjacencies.adjacencies_ordered_and_sorted(genomeA)
# Create start and target node
start_node = Node(adjacencies_genomeA)
def __init__(self, state=None):
self.state = state
get_chromosomes = Node_evolve.find_chromosomes(self, self.state)
self.linear_chromosomes = get_chromosomes[0]
self.circular_chromosomes = get_chromosomes[1]
get_adjacencies = Extremities_and_adjacencies()
randomized = 'Random weight ratios'
one_to_one = 'One to one weight ratios'
same_as_solution = 'Same as solution weight ratios'
#####################
number_of_simulations = 10000
number_of_sequence_blocks = 9
number_of_rearrangements = 8
type_of_weight_ratio = same_as_solution
results = []
t0 = time.time()
get_adjacencies = Extremities_and_adjacencies()
number_of_solutions_found = 0
get_adjacencies_and_genomes = Extremities_and_adjacencies()
for i in range(0, number_of_simulations):
#print('SIMULATION NUMBER: ', i, '*****************************************************************')
# print('num of sb, ', number_of_sequence_blocks)
target_genome = GenomeEvolver.create_target_genome(number_of_sequence_blocks)
evolving_genome = Evolve(target_genome)
rearrangement_series = evolving_genome.evolve_with_random_rearrangements(number_of_rearrangements)
reverse_the_series = GenomeEvolver.reverse_rearrangement_series(target_genome, rearrangement_series)
source_genome = reverse_the_series[1]
solution = reverse_the_series[2]
target_adjacencies = get_adjacencies_and_genomes.adjacencies_ordered_and_sorted(target_genome)
source_adjacencies = get_adjacencies_and_genomes.adjacencies_ordered_and_sorted(source_genome)
def transposition2(self):
get_adjacencies = Extremities_and_adjacencies()
# exsision and circularization
get_chromosomes = Node_evolve.find_chromosomes(self, self.state)
self.linear_chromosomes = get_chromosomes[0]
self.circular_chromosomes = get_chromosomes[1]
chrm_num = random.randint(0, len(self.linear_chromosomes) - 1)
# if it is a single gene or two gene chromosome, choose another one
while len(self.linear_chromosomes[chrm_num]) < 6:
chrm_num = random.randint(0, len(self.linear_chromosomes) - 1)
adjacencies = [
element for element in self.linear_chromosomes[chrm_num]
if type(element) is tuple
]
adj1_num = random.randint(0, len(adjacencies) - 1)
adj2_num = random.randint(0, len(adjacencies) - 1)
# To ensure single gene circular chromosomes are not formed because then there is only one adj so cut and not != join
while adj1_num == adj2_num or (int(adjacencies[adj2_num][0])) == int(
adjacencies[adj1_num][0]) or (int(
adjacencies[adj2_num][0])) == int(
adjacencies[adj1_num][1]) or (int(
adjacencies[adj2_num][1])) == int(
adjacencies[adj1_num][0]) or (int(
adjacencies[adj2_num][1])) == int(
adjacencies[adj1_num][1]):
adj2_num = random.randint(0, len(adjacencies) - 1)
adj1 = adjacencies[adj1_num]
adj2 = adjacencies[adj2_num]
if adj1[0] < adj2[1]:
new_adj1 = (adj1[0], adj2[1])
else:
new_adj1 = (adj2[1], adj1[0])
if adj1[1] < adj2[0]:
new_adj2 = (adj1[1], adj2[0])
else:
new_adj2 = (adj2[0], adj1[1])
# perfrom operation
self.state.remove(adj1)
self.state.remove(adj2)
self.state.append(new_adj1)
self.state.append(new_adj2)
get_chromosomes = Node_evolve.find_chromosomes(self, self.state)
self.linear_chromosomes = get_chromosomes[0]
self.circular_chromosomes = get_chromosomes[1]
# if the above does not result in the formation of a cicular chromosome join the adjacencies differently
# NTS you can actually take the above out and just start with this while loop...
while len(self.circular_chromosomes) == 0:
self.state.append(adj1)
self.state.append(adj2)
self.state.remove(new_adj1)
self.state.remove(new_adj2)
if adj1[0] < adj2[0]:
new_adj1 = (adj1[0], adj2[0])
else:
new_adj1 = (adj2[0], adj1[0])
if adj1[1] < adj2[1]:
new_adj2 = (adj1[1], adj2[1])
else:
new_adj2 = (adj2[1], adj1[1])
self.state.remove(adj1)
self.state.remove(adj2)
self.state.append(new_adj1)
self.state.append(new_adj2)
get_chromosomes = Node_evolve.find_chromosomes(self, self.state)
self.linear_chromosomes = get_chromosomes[0]
self.circular_chromosomes = get_chromosomes[1]
if new_adj1 in self.circular_chromosomes[0]:
join = new_adj1
excision = new_adj2
else:
join = new_adj2
excision = new_adj1
print('trp0: ', get_adjacencies.adjacencies_to_genome(self.state))
# decircularization and reinsertion
adj1_num = random.randint(0, len(self.circular_chromosomes[0]) - 1)
adj1 = self.circular_chromosomes[0][adj1_num]
while adj1 == join:
adj1_num = random.randint(0, len(self.circular_chromosomes[0]) - 1)
adj1 = self.circular_chromosomes[0][adj1_num]
chrm_num = random.randint(0, len(self.linear_chromosomes) - 1)
while len(self.linear_chromosomes[chrm_num]) == 2:
chrm_num = random.randint(0, len(self.linear_chromosomes) - 1)
if len(self.linear_chromosomes[chrm_num]) == 3:
adjacencies = [
element for element in self.linear_chromosomes[chrm_num]
if type(element) is tuple
]
adj2 = adjacencies[0]
else:
adjacencies = [
element for element in self.linear_chromosomes[chrm_num]
if type(element) is tuple
]
adj2_num = random.randint(0, len(adjacencies) - 1)
adj2 = adjacencies[adj2_num]
while adj2 == excision:
chrm_num = random.randint(0, len(self.linear_chromosomes) - 1)
# if it is a single gene chromosome, choose another chromosome
while len(self.linear_chromosomes[chrm_num]) == 2:
chrm_num = random.randint(0, len(self.linear_chromosomes) - 1)
if len(self.linear_chromosomes[chrm_num]) == 3:
adjacencies = [
element for element in self.linear_chromosomes[chrm_num]
if type(element) is tuple
]
adj2 = adjacencies[0]
else:
adjacencies = [
element for element in self.linear_chromosomes[chrm_num]
if type(element) is tuple
]
adj2_num = random.randint(0, len(adjacencies) - 1)
adj2 = adjacencies[adj2_num]
if adj1[0] < adj2[1]:
new_adj1 = (adj1[0], adj2[1])
else:
new_adj1 = (adj2[1], adj1[0])
if adj1[1] < adj2[0]:
new_adj2 = (adj1[1], adj2[0])
else:
new_adj2 = (adj2[0], adj1[1])
# perfrom operation
self.state.remove(adj1)
self.state.remove(adj2)
self.state.append(new_adj1)
self.state.append(new_adj2)
print('trp2: ', get_adjacencies.adjacencies_to_genome(self.state))
from Class_extremities_and_adjacencies import Extremities_and_adjacencies
from Class_Network_wrDCJ import Network
from Class_GraphTheory_weighted import GraphTheory
#genomeA = [[1, 2, 3, 5, 6, 4, 7, -8, 9]]
#genomeB = [[1, 2,3 ,4,5,6,7, 8, 9]]
#genomeA = [[1,-3,2,5,6,4,7]]
#genomeB = [[1, 2,3 ,4 , 5, 6, 7]]
#genomeA = [[1,-3,-2, 4, 5,6,9,7], [8, 10],[ 11, 12]]
#genomeB = [[1, 2,3 ,4 , 5, 6, 7], [8, 9, 10,11, 12]]
genomeA = [[1, 6, 7, 4, 5, 2, 3, -8, 9]]
genomeB = [[1, 2,3 ,4,5,6,7, 8, 9]]
#from genes to adjacencies
get_adjacencies = Extremities_and_adjacencies()
adjacencies_genomeA = get_adjacencies.adjacencies_ordered_and_sorted(genomeA)
adjacencies_genomeB = get_adjacencies.adjacencies_ordered_and_sorted(genomeB)
print('Adjacencies of the genomes: ')
print('Genome A: ', adjacencies_genomeA)
print('Genome B: ', adjacencies_genomeB)
print('____________________________________')
print()
print()
#Create start and target node
start_node = Node(adjacencies_genomeA)
def run(args):
genomeA_file = args.source_genome
genomeB_file = args.target_genome
weight_ratios_file = args.ratios
stdoutOrigin = sys.stdout
sys.stdout = open(args.output_file, 'w')
#outfile = open(args.output_file, 'w')
with open(genomeA_file) as csv:
line = [element.strip('\n').split(',') for element in csv]
genomeA = []
for element in line:
element = list(map(int, element))
genomeA.append(element)
with open(genomeB_file) as csv:
line = [element.strip('\n').split(',') for element in csv]
genomeB = []
for element in line:
element = list(map(int, element))
genomeB.append(element)
with open(weight_ratios_file) as csv:
line = [element.strip('\n').split(',') for element in csv]
weight_ratios = []
for element in line:
element = list(map(int, element))
weight_ratios.append(element)
get_adjacencies = Extremities_and_adjacencies()
adjacencies_genomeA = get_adjacencies.adjacencies_ordered_and_sorted(genomeA)
adjacencies_genomeB = get_adjacencies.adjacencies_ordered_and_sorted(genomeB)
#Create start and target node
start_node = Node(adjacencies_genomeA)
target_node = Node(adjacencies_genomeB)
hash_table = {}
hash_key_start = hash(str(start_node.state))
hash_key_target = hash(str(target_node.state))
hash_table.update({hash_key_start:start_node})
hash_table.update({hash_key_target:target_node})
#finding rearrangement weights
max_number = max(weight_ratios[0])
weights = []
for element in weight_ratios[0]:
if element == 0:
weights.append(max_number^2)
else:
weights.append(max_number/element)
New_Network_wrDCJ.build_hash_table(start_node, hash_table, adjacencies_genomeB, weights)
network = New_Network_wrDCJ.build_network(hash_table)
shortest_paths = (list(all_shortest_paths(network, start_node, target_node, weight='weight')))
j = 1
tot_b_trl = 0
tot_u_trl = 0
tot_inv = 0
tot_trp1 = 0
tot_trp2 = 0
tot_fus = 0
tot_fis = 0
Paths_state = []
Paths_state_weight = []
# print(shortest_paths[0][4].children_weights[2])
for path in shortest_paths:
path_state = []
path_state_weight = []
i = 0
while i < len(path):
current = path[i]
if i == 0:
operation_type = 'none, this is the source genome'
operation_weight = 'N/A'
operation = 'N/A'
else:
x = path[i - 1].children.index(current)
operation_type = path[i - 1].children_operations[x][1]
operation_weight = path[i - 1].children_weights[x]
operation = path[i - 1].children_operations[x][0]
adjacencies = current.state
genome = get_adjacencies.adjacencies_to_genome(adjacencies)
path_state_weight.append((genome, ((operation_type, operation), operation_weight)))
path_state.append((genome, (operation_type, operation)))
i += 1
Paths_state.append((path_state))
Paths_state_weight.append(path_state_weight)
for path in shortest_paths:
i = 0
b_trl = 0
u_trl = 0
inv = 0
trp1 = 0
trp2 = 0
fus = 0
fis = 0
while i < len(path):
current = path[i]
if i == 0:
pass
else:
x = path[i - 1].children.index(current)
operation_type = path[i - 1].children_operations[x][1]
if operation_type == 'b_trl':
b_trl += 1
elif operation_type == 'u_trl':
u_trl += 1
elif operation_type == 'inv':
inv += 1
elif operation_type == 'trp1':
trp1 += 1
elif operation_type == 'trp2':
trp2 += 1
elif operation_type == 'fus':
fus += 1
elif operation_type == 'fis':
fis += 1
i += 1
tot_b_trl += b_trl
tot_u_trl += u_trl
tot_inv += inv
tot_trp1 += trp1
tot_trp2 += trp2
tot_fus += fus
tot_fis += fis
j += 1
print('############################################################################################################')
print()
print('Source Genome: ', genomeA)
print('Target Genome: ', genomeB)
print()
print('Number of most parsimonious solutions: ', len(shortest_paths))
print()
print('Average number of operations per solution: ', float(tot_inv/len(shortest_paths))+float(tot_trp1/len(shortest_paths))+float(2*(tot_trp2/len(shortest_paths)))+float(tot_b_trl/len(shortest_paths))+float(tot_u_trl/len(shortest_paths))+float(tot_fis/len(shortest_paths))+float(tot_fus/len(shortest_paths)))
print()
print('Average number of each operation per solution:')
print('Inversions: ', float(tot_inv/len(shortest_paths)), ' Transpositions type 1: ', float(tot_trp1/len(shortest_paths)), ' Transpositions type 2: ', float(tot_trp2/len(shortest_paths)), ' Balanced translocations: ', float(tot_b_trl/len(shortest_paths)), ' Unbalanced translocations: ', float(tot_u_trl/len(shortest_paths)),
' Fusions: ', float(tot_fus/len(shortest_paths)),
' Fissions: ', float(tot_fis/len(shortest_paths)))
print()
print()
print('Solutions: ')
print()
path_counter = 1
for path in Paths_state:
print('Solution number ', path_counter)
for genome in path:
print(genome)
path_counter+=1
print()
print()
print('############################################################################################################')
sys.stdout.close()
sys.stdout=stdoutOrigin