def makeStarting(tips,RF_norm_start,name,RF_norm_cloud,cloud_size):
#Choose starting trees
print(name)
print("Making starting trees...")
# Create random tree. Average branch length of 1. No variation in branch length. Don't print out tree structure
t1 = readTree.rand_tree(tips=tips,brl_avg=1,brl_std=None,verbose='F')
# Calculate number of NNI moves based on desired normalized RF distance.
RF_max = 2*(tips-2)
NNI_moves_start = int((RF_max * RF_norm_start)/2)
if NNI_moves_start == 0:
NNI_moves_start = 1
NNI_moves_cloud = int((RF_max * RF_norm_cloud)/2)
# Create second starting tree
t2 = readTree.NNI_mult_moves(in_tree=t1,num_moves=NNI_moves_start,node_choice='random',no_dup_start_tree='F', req_min_RF=RF_norm_start)
# Write out tree files
readTree.write_single_tree(t1,'%s_starting_tree_1.tree' % name)
readTree.write_single_tree(t2,'%s_starting_tree_2.tree' % name)
# Write out log file
# Calculate emperical distance between two start trees as a gut check
RF_emp = readTree.rf_unweighted(t1,t2,normalized='T')[1]
# write to log file
with open('%s.log' % name, "w") as log_file:
line1 = "File name: "+str(name)
line1b = "Tips: "+str(tips)+", Trees per cloud: "+str(cloud_size)
line2 = "Starting trees - RF_input: "+str(RF_norm_start)+", RF_calc: "+str(RF_emp)+", NNI_moves: "+str(NNI_moves_start)
line3 = "Cloud of trees - RF_input: "+str(RF_norm_cloud)+", NNI_moves: "+str(NNI_moves_cloud)
log_file.write("%s\n%s\n%s" % (line1, line2, line3))
# Pass tree files to next function
return t1,t2
def makeStarting(tips, RF_norm_start, name, RF_norm_cloud, cloud_size):
#Choose starting trees
print(name)
print("Making starting trees...")
# Create random tree. Average branch length of 1. No variation in branch length. Don't print out tree structure
t1 = readTree.rand_tree(tips=tips, brl_avg=1, brl_std=None, verbose='F')
# Calculate number of NNI moves based on desired normalized RF distance.
RF_max = 2 * (tips - 2)
NNI_moves_start = int((RF_max * RF_norm_start) / 2)
if NNI_moves_start == 0:
NNI_moves_start = 1
# Calc cloud for log file
NNI_moves_cloud = int((RF_max * RF_norm_cloud) / 2)
# Create second starting tree
t2 = readTree.NNI_mult_moves(in_tree=t1,
num_moves=NNI_moves_start,
node_choice='random',
no_dup_start_tree='F',
req_min_RF=RF_norm_start)
readTree.rf_unweighted(t1, t2, "T")
t3 = readTree.NNI_mult_moves(in_tree=t1,
num_moves=NNI_moves_start,
node_choice='random',
no_dup_start_tree='F',
req_min_RF=RF_norm_start)
d12 = readTree.rf_unweighted(t1, t2, "T")[1]
d23 = readTree.rf_unweighted(t2, t3, "T")[1]
d13 = readTree.rf_unweighted(t1, t3, "T")[1]
print("first t2 to t3 RF" + str(d23))
#t1 and t2 are guarenteed to be RF start apart, same with t2 and t3. So we only need to check t1 and t3.
while d23 != RF_norm_start:
print("redoing third start tree")
t3 = readTree.NNI_mult_moves(in_tree=t1,
num_moves=NNI_moves_start,
node_choice='random',
no_dup_start_tree='F',
req_min_RF=RF_norm_start)
d23 = readTree.rf_unweighted(t2, t3, "T")[1]
print(d23)
# Write out tree files
readTree.write_single_tree(t1, '%s_starting_tree_1.tree' % name)
readTree.write_single_tree(t2, '%s_starting_tree_2.tree' % name)
readTree.write_single_tree(t3, '%s_starting_tree_3.tree' % name)
# Write out log file
rfs = str(d12) + ", " + str(d13) + ", " + str(d23)
with open('%s.log' % name, "w") as log_file:
line1 = "File name: " + str(name)
line1b = "Tips: " + str(tips) + ", Trees per cloud: " + str(cloud_size)
line2 = "Starting trees - RF_input: " + str(
RF_norm_start) + ", RF_calc: " + str(rfs) + ", NNI_moves: " + str(
NNI_moves_start)
line3 = "Cloud of trees - RF_input: " + str(
RF_norm_cloud) + ", NNI_moves: " + str(NNI_moves_cloud)
log_file.write("%s\n%s\n%s" % (line1, line2, line3))
# Pass tree files to next function
return t1, t2, t3
num_out_trees=c_size,
num_nni_moves=NNI_moves_cloud,
out='list')
cluster2 = readTree.NNI_mult_trees(in_tree=t2,
num_out_trees=c_size,
num_nni_moves=NNI_moves_cloud,
out='list')
# Make a nexus file with starting trees and cloud trees
readTree.list_to_out(cluster1, cluster2, '%s_cloud.tree' % name)
############################################################
# Print info to log file
############################################################
print("Calculating stats on trees...")
# Calculate emperical distance between two start trees
RF_emp = readTree.rf_unweighted(t1, t2, normalized='T')[1]
# Calculate average density of each cloud
RF_cloud1 = readTree.cluster_density_avg(in_file='%s_cloud.tree' % name,
NNI_trees=int(cloud_size) - 1,
starting_tree_number=0)
RF_cloud2 = readTree.cluster_density_avg(in_file='%s_cloud.tree' % name,
NNI_trees=int(cloud_size) - 1,
starting_tree_number=cloud_size)
# write to log file
with open('%s.log' % name, "w") as log_file:
line1 = "File name: " + str(name)
line1b = "Tips: " + str(tips) + ", Trees per cloud: " + str(cloud_size)
line2 = "Starting trees - RF_input: " + str(
RF_norm_start) + ", RF_calc: " + str(RF_emp) + ", NNI_moves: " + str(
NNI_moves_start)
line3 = "Cloud of trees - RF_input: " + str(
def makeStarting(tips, RF_norm_start_far, RF_norm_start, name, RF_norm_cloud,
cloud_size):
#Choose starting trees
print(name)
print("Making starting trees...")
#Round the things
RF_norm_start_far = round(RF_norm_start_far, 2)
RF_norm_start = round(RF_norm_start, 2)
# Calculate number of NNI moves based on desired normalized RF distance.
RF_max = 2 * (tips - 2)
NNI_moves_start = int((RF_max * RF_norm_start) / 2)
if NNI_moves_start == 0:
NNI_moves_start = 1
NNI_moves_start_far = int((RF_max * RF_norm_start_far) / 2)
print("NNI: " + str(NNI_moves_start) + "NNI far: " +
str(NNI_moves_start_far))
# Calc cloud for log file
NNI_moves_cloud = int((RF_max * RF_norm_cloud) / 2)
# Create random tree. Average branch length of 1. No variation in branch length. Don't print out tree structure
t1 = readTree.rand_tree(tips=tips, brl_avg=1, brl_std=None, verbose='F')
# Create second starting tree
t2 = readTree.NNI_mult_moves(in_tree=t1,
num_moves=NNI_moves_start,
node_choice='random',
no_dup_start_tree='F',
req_min_RF=RF_norm_start)
t3 = readTree.NNI_mult_moves(in_tree=t1,
num_moves=NNI_moves_start,
node_choice='random',
no_dup_start_tree='F',
req_min_RF=RF_norm_start)
d12 = readTree.rf_unweighted(t1, t2, "T")[1]
d23 = readTree.rf_unweighted(t2, t3, "T")[1]
d13 = readTree.rf_unweighted(t1, t3, "T")[1]
print("first t1 to t2 RF - " + str(round(d12, 2)))
print("first t1 to t3 RF - " + str(round(d13, 2)))
print("first t2 to t3 RF - " + str(round(d23, 2)))
#t1 and t2 are guarenteed to be RF start apart, same with t2 and t3. So we only need to check t1 and t3.
while round(d23, 2) != RF_norm_start:
print("redoing third start tree - " + name)
print(round(d23, 2), RF_norm_start)
t3 = readTree.NNI_mult_moves(in_tree=t1,
num_moves=NNI_moves_start,
node_choice='random',
no_dup_start_tree='F',
req_min_RF=RF_norm_start)
d23 = readTree.rf_unweighted(t2, t3, "T")[1]
print("t1 to t2 RF - " + str(round(d12, 2)))
print("t1 to t3 RF - " + str(round(d13, 2)))
print("t2 to t3 RF - " + str(round(d23, 2)))
# do it all again with another starting tree
t4 = readTree.NNI_mult_moves(in_tree=t1,
num_moves=NNI_moves_start_far,
node_choice='random',
no_dup_start_tree='F',
req_min_RF=RF_norm_start_far)
# now we need to check d24 and d34
d14 = readTree.rf_unweighted(t1, t4, "T")[1]
d24 = readTree.rf_unweighted(t2, t4, "T")[1]
d34 = readTree.rf_unweighted(t3, t4, "T")[1]
print("first t1 to t4 RF - " + str(round(d14, 2)))
print("first t2 to t4 RF - " + str(round(d24, 2)))
print("first t3 to t4 RF - " + str(round(d34, 2)))
while round(d24, 2) != RF_norm_start_far or round(d34,
2) != RF_norm_start_far:
print("redoing fourth start tree, Far Tree - " + name)
print(round(d24, 2), round(d34, 2), RF_norm_start_far)
t4 = readTree.NNI_mult_moves(in_tree=t1,
num_moves=NNI_moves_start_far,
node_choice='random',
no_dup_start_tree='F',
req_min_RF=RF_norm_start_far)
d24 = readTree.rf_unweighted(t2, t4, "T")[1]
d34 = readTree.rf_unweighted(t3, t4, "T")[1]
# Write out tree files
readTree.write_single_tree(t1, '%s_starting_tree_1.tree' % name)
readTree.write_single_tree(t2, '%s_starting_tree_2.tree' % name)
readTree.write_single_tree(t3, '%s_starting_tree_3.tree' % name)
readTree.write_single_tree(t4, '%s_starting_tree_4.tree' % name)
# Write out log file
rfs = str(d12) + ", " + str(d13) + ", " + str(d23) + ", " + str(
d14) + ", " + str(d24) + ", " + str(d34)
with open('%s.log' % name, "w") as log_file:
line1 = "File name: " + str(name)
line1b = "Tips: " + str(tips) + ", Trees per cloud: " + str(cloud_size)
line2 = "Starting trees - RF_input: " + str(
RF_norm_start) + "RF_input_far: " + str(
RF_norm_start_far) + ", RF_calc: " + str(
rfs) + ", NNI_moves: " + str(
NNI_moves_start) + ", NNI_moves_far: " + str(
NNI_moves_start_far)
line3 = "Cloud of trees - RF_input: " + str(
RF_norm_cloud) + ", NNI_moves: " + str(NNI_moves_cloud)
log_file.write("%s\n%s\n%s" % (line1, line2, line3))
# Pass tree files to next function
return t1, t2, t3, t4
############################## # Make moves ############################## # Print newick string in_tree_object.newick(in_tree_object.root) # Make NNI move new_tree_object = readTree.NNI(orig_tree=in_tree_object,node_choice='exponential') # Make multiple moves on a single tree new_tree_object = readTree.NNI_mult_moves(in_tree=in_tree,num_moves=1,node_choice='exponential',no_dup_start_tree='T') # Make multiple trees, each with one NNI move from starting tree. Output as nexus file. readTree.NNI_mult_trees(in_tree=in_tree,num_out_trees=10,num_nni_moves=2,out_file='outFile2.t',node_choice='random',no_dup_start_tree='T') ############################## # Look at and compare trees ############################## # Print newick string new_tree_object.newick(new_tree_object.root) # View tree in terminal along with newick string readTree.view_phylo(new_tree_object) # Check that all trees have the right number of moves from start tree readTree.compare_tree_file(in_file='outFile1.t',total_trees=11,distance_metric="uRF") # Compare RF distance between two trees. readTree.rf_unweighted(in_tree_object,new_tree_object,normalized='T')
