def infer_tree(self, raxml_time_limit=1.0):
import tree_infer
tree_fname = tree_infer.main(self.viruses, raxml_time_limit,
config['outgroup'])
delimit_newick(tree_fname, "temp.newick")
self.tree = dendropy.Tree.get_from_path("temp.newick", "newick")
os.remove('temp.newick')
def infer_tree(self, raxml_time_limit):
'''
builds a tree from the alignment using fasttree and RAxML. raxml runs for
raxml_time_limit and is terminated thereafter. raxml_time_limit can be 0.
'''
self.make_run_dir()
os.chdir(self.run_dir)
AlignIO.write(self.viruses, 'temp.fasta', 'fasta')
print "Building initial tree with FastTree"
os.system(
"/Users/yujiazhou/Documents/nextflu/fasttree -gtr -nt -gamma -nosupport -mlacc 2 -slownni temp.fasta > initial_tree.newick"
)
self.tree = dendropy.Tree.get_from_string(
delimit_newick('initial_tree.newick'),
'newick',
rooting='force-rooted')
self.tree.resolve_polytomies()
self.tree.write_to_path("initial_tree.newick", "newick")
AlignIO.write(self.viruses, "temp.phyx", "phylip-relaxed")
if raxml_time_limit > 0:
print "RAxML tree optimization with time limit " + str(
raxml_time_limit) + " hours"
# using exec to be able to kill process
end_time = time.time() + int(raxml_time_limit * 3600)
process = subprocess.Popen(
"exec /Users/yujiazhou/Documents/nextflu/raxmlHPC-AVX-v8/raxml -f d -T 6 -j -s temp.phyx -n topology -c 25 -m GTRCAT -p 344312987 -t initial_tree.newick",
shell=True)
while (time.time() < end_time):
if os.path.isfile('RAxML_result.topology'):
break
time.sleep(10)
process.terminate()
checkpoint_files = glob.glob("RAxML_checkpoint*")
if os.path.isfile('RAxML_result.topology'):
checkpoint_files.append('RAxML_result.topology')
if len(checkpoint_files) > 0:
last_tree_file = checkpoint_files[-1]
shutil.copy(last_tree_file, 'raxml_tree.newick')
else:
shutil.copy("initial_tree.newick", 'raxml_tree.newick')
else:
shutil.copy("initial_tree.newick", 'raxml_tree.newick')
print "RAxML branch length optimization and rooting"
os.system(
"/Users/yujiazhou/Documents/nextflu/raxmlHPC-AVX-v8/raxml -f e -T 6 -s temp.phyx -n branches -c 25 -m GTRGAMMA -p 344312987 -t raxml_tree.newick -o "
+ self.outgroup['strain'])
out_fname = "tree_infer.newick"
shutil.copy('RAxML_result.branches', out_fname)
Phylo.write(Phylo.read(out_fname, 'newick'), 'temp.newick', 'newick')
self.tree = dendropy.Tree.get_from_string(delimit_newick(out_fname),
'newick',
rooting='force-rooted')
os.chdir('..')
self.remove_run_dir()
def main(viruses, raxml_time_limit, outgroup):
print "--- Tree infer at " + time.strftime("%H:%M:%S") + " ---"
cleanup()
write_fasta(viruses, 'temp.fasta')
print "Building initial tree with FastTree"
os.system(
"./fasttree -gtr -nt -gamma -nosupport -mlacc 2 -slownni temp.fasta > initial_tree.newick"
)
delimit_newick("initial_tree.newick", "temp.newick")
tree = dendropy.Tree.get_from_path("temp.newick", "newick")
tree.resolve_polytomies()
tree.write_to_path("initial_tree.newick", "newick")
os.system("seqmagick convert temp.fasta temp.phyx")
if raxml_time_limit > 0:
print "RAxML tree optimization with time limit " + str(
raxml_time_limit) + " hours"
# using exec to be able to kill process
end_time = time.time() + int(raxml_time_limit * 3600)
process = subprocess.Popen(
"exec raxml -f d -T 6 -j -s temp.phyx -n topology -c 25 -m GTRCAT -p 344312987 -t initial_tree.newick",
shell=True)
while (time.time() < end_time):
if os.path.isfile('RAxML_result.topology'):
break
time.sleep(10)
process.terminate()
checkpoint_files = [file for file in glob.glob("RAxML_checkpoint*")]
if os.path.isfile('RAxML_result.topology'):
checkpoint_files.append('RAxML_result.topology')
if len(checkpoint_files) > 0:
last_tree_file = checkpoint_files[-1]
shutil.copy(last_tree_file, 'raxml_tree.newick')
else:
shutil.copy("initial_tree.newick", 'raxml_tree.newick')
else:
shutil.copy("initial_tree.newick", 'raxml_tree.newick')
print "RAxML branch length optimization and rooting"
os.system(
"raxml -f e -T 6 -s temp.phyx -n branches -c 25 -m GTRGAMMA -p 344312987 -t raxml_tree.newick -o "
+ outgroup)
out_fname = "data/tree_infer.newick"
os.rename('RAxML_result.branches', out_fname)
cleanup()
return out_fname
def main(viruses, raxml_time_limit, outgroup):
print "--- Tree infer at " + time.strftime("%H:%M:%S") + " ---"
cleanup()
write_fasta(viruses, "temp.fasta")
print "Building initial tree with FastTree"
os.system("fasttree -gtr -nt -gamma -nosupport -mlacc 2 -slownni temp.fasta > initial_tree.newick")
delimit_newick("initial_tree.newick", "temp.newick")
tree = dendropy.Tree.get_from_path("temp.newick", "newick")
tree.resolve_polytomies()
tree.write_to_path("initial_tree.newick", "newick")
os.system("seqmagick convert temp.fasta temp.phyx")
if raxml_time_limit > 0:
print "RAxML tree optimization with time limit " + str(raxml_time_limit) + " hours"
# using exec to be able to kill process
end_time = time.time() + int(raxml_time_limit * 3600)
process = subprocess.Popen(
"exec raxml -f d -T 6 -j -s temp.phyx -n topology -c 25 -m GTRCAT -p 344312987 -t initial_tree.newick",
shell=True,
)
while time.time() < end_time:
if os.path.isfile("RAxML_result.topology"):
break
time.sleep(10)
process.terminate()
checkpoint_files = [file for file in glob.glob("RAxML_checkpoint*")]
if os.path.isfile("RAxML_result.topology"):
checkpoint_files.append("RAxML_result.topology")
if len(checkpoint_files) > 0:
last_tree_file = checkpoint_files[-1]
shutil.copy(last_tree_file, "raxml_tree.newick")
else:
shutil.copy("initial_tree.newick", "raxml_tree.newick")
else:
shutil.copy("initial_tree.newick", "raxml_tree.newick")
print "RAxML branch length optimization and rooting"
os.system(
"raxml -f e -T 6 -s temp.phyx -n branches -c 25 -m GTRGAMMA -p 344312987 -t raxml_tree.newick -o " + outgroup
)
out_fname = "data/tree_infer.newick"
os.rename("RAxML_result.branches", out_fname)
cleanup()
return out_fname
def infer_tree(self, raxml_time_limit):
'''
builds a tree from the alignment using fasttree and RAxML. raxml runs for
raxml_time_limit and is terminated thereafter. raxml_time_limit can be 0.
'''
self.make_run_dir()
os.chdir(self.run_dir)
AlignIO.write(self.viruses, 'temp.fasta', 'fasta')
print "Building initial tree with FastTree"
os.system("fasttree -gtr -nt -gamma -nosupport -mlacc 2 -slownni temp.fasta > initial_tree.newick")
self.tree = dendropy.Tree.get_from_string(delimit_newick('initial_tree.newick'),'newick', as_rooted=True)
self.tree.resolve_polytomies()
self.tree.write_to_path("initial_tree.newick", "newick")
AlignIO.write(self.viruses,"temp.phyx", "phylip-relaxed")
if raxml_time_limit>0:
print "RAxML tree optimization with time limit " + str(raxml_time_limit) + " hours"
# using exec to be able to kill process
end_time = time.time() + int(raxml_time_limit*3600)
process = subprocess.Popen("exec raxml -f d -T 6 -j -s temp.phyx -n topology -c 25 -m GTRCAT -p 344312987 -t initial_tree.newick", shell=True)
while (time.time() < end_time):
if os.path.isfile('RAxML_result.topology'):
break
time.sleep(10)
process.terminate()
checkpoint_files = glob.glob("RAxML_checkpoint*")
if os.path.isfile('RAxML_result.topology'):
checkpoint_files.append('RAxML_result.topology')
if len(checkpoint_files) > 0:
last_tree_file = checkpoint_files[-1]
shutil.copy(last_tree_file, 'raxml_tree.newick')
else:
shutil.copy("initial_tree.newick", 'raxml_tree.newick')
else:
shutil.copy("initial_tree.newick", 'raxml_tree.newick')
print "RAxML branch length optimization and rooting"
os.system("raxml -f e -T 6 -s temp.phyx -n branches -c 25 -m GTRGAMMA -p 344312987 -t raxml_tree.newick -o " + self.outgroup['strain'])
out_fname = "tree_infer.newick"
shutil.copy('RAxML_result.branches', out_fname)
Phylo.write(Phylo.read(out_fname, 'newick'),'temp.newick','newick')
self.tree = dendropy.Tree.get_from_string(delimit_newick(out_fname), 'newick', as_rooted=True)
os.chdir('..')
self.remove_run_dir()
def infer_tree(self, raxml_time_limit = 1.0):
import tree_infer
tree_fname = tree_infer.main(self.viruses, raxml_time_limit, config['outgroup'])
delimit_newick(tree_fname, "temp.newick")
self.tree = dendropy.Tree.get_from_path("temp.newick", "newick")
os.remove('temp.newick')
def infer_tree(self, raxml_time_limit):
'''
builds a tree from the alignment using fasttree and RAxML. raxml runs for
raxml_time_limit and is terminated thereafter. raxml_time_limit can be 0.
'''
self.make_run_dir()
os.chdir(self.run_dir)
AlignIO.write(self.viruses, 'temp.fasta', 'fasta')
print "Building initial tree with FastTree"
os.system("fasttree -gtr -nt -gamma -nosupport temp.fasta 1> initial_tree.newick 2>fasttree.out")
self.tree = dendropy.Tree.get_from_string(delimit_newick('initial_tree.newick'),'newick', as_rooted=True)
self.tree.resolve_polytomies()
self.tree.write_to_path("initial_tree.newick", "newick")
AlignIO.write(self.viruses,"temp.phyx", "phylip-relaxed")
if raxml_time_limit>0:
print "RAxML tree optimization with time limit " + str(raxml_time_limit) + " hours"
# using exec to be able to kill process
end_time = time.time() + int(raxml_time_limit*3600)
process = subprocess.Popen("exec raxmlHPC -f d -T "+str(self.nthreads) + " -j -s temp.phyx -n topology -c 25 -m GTRCAT -p 344312987 -t initial_tree.newick >raxml.out", shell=True)
while (time.time() < end_time):
if os.path.isfile('RAxML_result.topology'):
break
time.sleep(10)
process.terminate()
checkpoint_files = [file for file in glob.glob("RAxML_checkpoint*")]
if os.path.isfile('RAxML_result.topology'):
checkpoint_files.append('RAxML_result.topology')
if len(checkpoint_files) > 0:
last_tree_file = checkpoint_files[-1]
shutil.copy(last_tree_file, 'raxml_tree.newick')
else:
shutil.copy("initial_tree.newick", 'raxml_tree.newick')
else:
shutil.copy("initial_tree.newick", 'raxml_tree.newick')
if raxml_time_limit>0:
print "RAxML branch length optimization and rooting"
os.system("raxmlHPC -f e -T "+str(self.nthreads) + " -s temp.phyx -n branches -c 25 -m GTRGAMMA -p 344312987 -t raxml_tree.newick -o " + self.outgroup['strain']+ ' >raxml2.out')
raxml_rooted=True
else:
shutil.copy('raxml_tree.newick', 'RAxML_result.branches')
raxml_rooted=False
out_fname = "tree_infer.newick"
shutil.copy('RAxML_result.branches', out_fname)
if not raxml_rooted:
with open(out_fname) as ofile:
tstr = "".join([x.strip() for x in ofile])
out_fname = out_fname+'_fixed'
if tstr.startswith('[&R]'):
with open(out_fname,'w') as ofile:
ofile.write(tstr[4:]+'\n')
T = Phylo.read(out_fname, 'newick')
try:
outgroup_clade = [c for c in T.get_terminals() if c.name == self.outgroup['strain']][0]
except:
print("Can't find outgroup in tree -- midpoint_rooting")
self.midpoint_rooting = True
else:
T = Phylo.read(out_fname, 'newick')
Phylo.write(T,'temp.newick','newick')
self.tree = dendropy.Tree.get_from_string(delimit_newick(out_fname), 'newick', rooting="force-rooted")
os.chdir('..')
self.remove_run_dir()
if self.midpoint_rooting:
self.tree.reroot_at_midpoint()
