def add(alignment, sequence, timeout, logger, wd, threads):
"""Align sequence(s) to an alignment using mafft (external
program)"""
alignment_file = "alignment_in.fasta"
sequence_file = "sequence_in.fasta"
output_file = "alignment_out.fasta" + ".fasta"
command_line = "{0} --auto --thread {1} --add {2} {3} > {4}".format(
mafft, threads, sequence_file, alignment_file, output_file
)
with open(os.path.join(wd, sequence_file), "w") as file:
SeqIO.write(sequence, file, "fasta")
with open(os.path.join(wd, alignment_file), "w") as file:
AlignIO.write(alignment, file, "fasta")
pipe = TerminationPipe(command_line, timeout=timeout, cwd=wd)
pipe.run()
os.remove(os.path.join(wd, alignment_file))
os.remove(os.path.join(wd, sequence_file))
if not pipe.failure:
try:
res = AlignIO.read(os.path.join(wd, output_file), "fasta")
except:
logger.info(pipe.output)
raise MafftError()
else:
os.remove(os.path.join(wd, output_file))
else:
logger.debug(".... add timeout ....")
return genNonAlignment(len(alignment) + 1, len(alignment.get_alignment_length()))
return res
def align(command, sequences, timeout, logger, wd, threads):
"""Adapted pG function: Align sequences using mafft (external
program)"""
input_file = "sequences_in.fasta"
output_file = "alignment_out.fasta"
command_line = "{0} --thread {1} {2} > {3}".format(command, threads, input_file, output_file)
with open(os.path.join(wd, input_file), "w") as file:
SeqIO.write(sequences, file, "fasta")
logger.debug(command_line)
pipe = TerminationPipe(command_line, timeout=timeout, cwd=wd)
pipe.run()
os.remove(os.path.join(wd, input_file))
if not pipe.failure:
try:
res = AlignIO.read(os.path.join(wd, output_file), "fasta")
except:
logger.info(pipe.output)
raise MafftError()
else:
os.remove(os.path.join(wd, output_file))
else:
# if pipe.failure, runtime error, return non-alignment
logger.debug(".... align timeout ....")
return genNonAlignment(len(sequences), len(sequences[0]))
return res
def align(command, sequences, timeout, logger, wd, threads):
"""Adapted pG function: Align sequences using mafft (external
program)"""
input_file = "sequences_in.fasta"
output_file = "alignment_out.fasta"
command_line = '{0} --thread {1} {2} > {3}'.format(command, threads,
input_file, output_file)
with open(os.path.join(wd, input_file), "w") as file:
SeqIO.write(sequences, file, "fasta")
logger.debug(command_line)
pipe = TerminationPipe(command_line, timeout=timeout, cwd=wd)
pipe.run()
os.remove(os.path.join(wd, input_file))
if not pipe.failure:
try:
res = AlignIO.read(os.path.join(wd, output_file), 'fasta')
except:
logger.info(pipe.output)
raise MafftError()
else:
os.remove(os.path.join(wd, output_file))
else:
# if pipe.failure, runtime error, return non-alignment
logger.debug('.... align timeout ....')
return genNonAlignment(len(sequences), len(sequences[0]))
return res
def add(alignment, sequence, timeout, logger, wd, threads):
"""Align sequence(s) to an alignment using mafft (external
program)"""
alignment_file = "alignment_in.fasta"
sequence_file = "sequence_in.fasta"
output_file = "alignment_out.fasta" + '.fasta'
command_line = '{0} --auto --thread {1} --add {2} {3} > {4}'.\
format(mafft, threads, sequence_file, alignment_file,
output_file)
with open(os.path.join(wd, sequence_file), "w") as file:
SeqIO.write(sequence, file, "fasta")
with open(os.path.join(wd, alignment_file), "w") as file:
AlignIO.write(alignment, file, "fasta")
pipe = TerminationPipe(command_line, timeout=timeout, cwd=wd)
pipe.run()
os.remove(os.path.join(wd, alignment_file))
os.remove(os.path.join(wd, sequence_file))
if not pipe.failure:
try:
res = AlignIO.read(os.path.join(wd, output_file), 'fasta')
except:
logger.info(pipe.output)
raise MafftError()
else:
os.remove(os.path.join(wd, output_file))
else:
logger.debug('.... add timeout ....')
return genNonAlignment(
len(alignment) + 1, len(alignment.get_alignment_length()))
return res
def RAxML(alignment, wd, logger, threads, outgroup=None, partitions=None,
constraint=None, timeout=999999999):
"""Adapted pG function: Generate phylogeny from alignment using
RAxML (external program)."""
# TODO: too complex, consider breaking up
input_file = 'phylogeny_in.phylip'
output_file = 'phylogeny_out'
file_line = ' -s ' + input_file + ' -n ' + output_file
options = ' -p ' + str(random.randint(0, 10000000))
# only add -T arg if pthreads version
if pthreads:
options += ' -T ' + str(threads)
if outgroup:
options += ' -o ' + outgroup
with open(os.path.join(wd, input_file), "w") as file:
AlignIO.write(alignment, file, "phylip-relaxed")
# only use GTRCAT for more than 100 taxa (ref RAxML manual)
if len(alignment) > 100:
dnamodel = ' -m GTRCAT'
else:
dnamodel = ' -m GTRGAMMA'
if partitions:
options += partitions
if constraint:
options += constraint
command_line = raxml + file_line + dnamodel + options
logger.debug(command_line)
pipe = TerminationPipe(command_line, silent=True, cwd=wd)
pipe.run()
if not pipe.failure:
try:
with open(os.path.join(wd, 'RAxML_bestTree.' + output_file), "r") \
as file:
tree = Phylo.read(file, "newick")
except IOError:
return None
finally:
if constraint:
os.remove(os.path.join(wd, 'constraint.tre'))
if partitions:
os.remove(os.path.join(wd, "partitions.txt"))
os.remove(os.path.join(wd, input_file))
all_files = os.listdir(wd)
for each in all_files:
if re.search("(RAxML)", each):
os.remove(os.path.join(wd, each))
if re.search("\.reduced$", each):
os.remove(os.path.join(wd, each))
return tree
else:
raise RuntimeError()
