def pll_task(alignment_file, partition_string, guidetree=None, threads=1, seed=PLL_RANDOM_SEED, frequencies=None,
write_to_file=None):
guidetree = True if guidetree is None else guidetree
instance = pll(alignment_file, partition_string, guidetree, threads, seed)
if frequencies is not None and len(frequencies) == instance.get_number_of_partitions():
for i in range(len(frequencies)):
instance.set_frequencies(frequencies[i], i, False)
instance.optimise_tree_search(True)
result = pll_to_dict(instance)
if write_to_file is not None: # attempt to write to file specified by write_to_file
try:
parameters = Parameters()
parameters.construct_from_dict(result)
with fileIO.fwriter(write_to_file, gz=True) as fl:
parameters.write(fl)
except:
pass # fail silently
return result
def pll_task(alignment_file,
partition_string,
guidetree=None,
threads=1,
seed=PLL_RANDOM_SEED,
frequencies=None,
write_to_file=None):
guidetree = True if guidetree is None else guidetree
instance = pll(alignment_file, partition_string, guidetree, threads, seed)
if frequencies is not None and len(
frequencies) == instance.get_number_of_partitions():
for i in range(len(frequencies)):
instance.set_frequencies(frequencies[i], i, False)
instance.optimise_tree_search(True)
result = pll_to_dict(instance)
if write_to_file is not None: # attempt to write to file specified by write_to_file
try:
parameters = Parameters()
parameters.construct_from_dict(result)
with fileIO.fwriter(write_to_file, gz=True) as fl:
parameters.write(fl)
except:
pass # fail silently
return result
return None
try:
ac, ag, at, cg, ct, gt = (float(x) for x in re.search(r'GTR rates\(ac ag at cg ct gt\) ([0-9.]+)\s+([0-9.]+)\s+([0-9.]+)\s+([0-9.]+)\s+([0-9.]+)\s+([0-9.]+)', s).groups())
except AttributeError:
warnings.warn('Couldn\'t parse substitution rates from {}'.format(filename))
return None
return {
'loglk': loglk,
'alpha': alpha,
'freqs': np.array([a, c, g, t]),
'rates': np.array([ [0 , ac, ag, at],
[ac, 0 , cg, ct],
[ag, cg, 0 , gt],
[at, ct, gt, 0 ] ])
}
if __name__ == '__main__':
import argparse
parser = argparse.ArgumentParser()
parser.add_argument('-f', '--filename', type=str)
parser.add_argument('-o', '--output', type=str)
args = parser.parse_args()
import json
result = run_raxml_fast(args.filename)
with fileIO.fwriter(args.output) as fl:
json.dump(result, fl)
def write_concats(collection, ps, path, index, logger, method, fastest):
concat_dir = get_dirs(path, index)['concdir']
if not os.path.exists(concat_dir): os.mkdir(concat_dir)
# First concat is all records, no clustering
outf = os.path.join(concat_dir, '1cl0.phy')
outr = os.path.join(concat_dir, '1cl0.json')
if not os.path.exists(outf):
conc = collection.concatenate(range(len(collection)))
al = conc.alignment
al.write_alignment(outf, 'phylip', True)
AlignIO.convert(outf, 'phylip-relaxed', '{}_'.format(outf), 'phylip-relaxed')
os.system('mv {} {}'.format('{}_'.format(outf), outf))
if not os.path.exists(outr):
al = treeCl.alignment.Alignment(outf, 'phylip', True)
if al.is_dna():
if method == 'raxml':
logger.info('Calculating fast raxml tree for dna alignment 1cl0.phy')
result = run_raxml_fast(outf, fastest)
else:
logger.info('Calculating FastTree tree for dna alignment 1cl0.phy')
result = run_fasttree(outf, True)
else:
if method == 'raxml':
logger.info('Calculating fast raxml tree for protein alignment 1cl0.phy')
result = run_raxml_fast(outf, 'PROTGAMMALG', fastest)
else:
logger.info('Calculating FastTree tree for protein alignment 1cl0.phy')
result = run_fasttree(outf, False)
with fileIO.fwriter(outr) as fl:
json.dump(result, fl)
# Now do rest of concats
for ngrp, p in enumerate(ps, start=2):
for i, grp in enumerate(p.get_membership()):
outf = os.path.join(concat_dir, '{}cl{}.phy'.format(ngrp, i))
outr = os.path.join(concat_dir, '{}cl{}.json'.format(ngrp, i))
if not os.path.exists(outf):
conc = collection.concatenate(grp)
al = conc.alignment
al.write_alignment(outf, 'phylip', True)
AlignIO.convert(outf, 'phylip-relaxed', '{}_'.format(outf), 'phylip-relaxed')
os.system('mv {} {}'.format('{}_'.format(outf), outf))
if not os.path.exists(outr):
al = treeCl.alignment.Alignment(outf, 'phylip', True)
if al.is_dna():
if method == 'raxml':
logger.info('Calculating fast raxml tree for dna alignment {}cl{}.phy'.format(ngrp, i))
result = run_raxml_fast(outf, fastest)
else:
logger.info('Calculating FastTree tree for dna alignment {}cl{}.phy'.format(ngrp, i))
result = run_fasttree(outf, True)
else:
if method == 'raxml':
logger.info('Calculating fast raxml tree for protein alignment {}cl{}.phy'.format(ngrp, i))
result = run_raxml_fast(outf, 'PROTGAMMALG', fastest)
else:
logger.info('Calculating FastTree tree for protein alignment {}cl{}.phy'.format(ngrp, i))
result = run_fasttree(outf, False)
with fileIO.fwriter(outr) as fl:
json.dump(result, fl)
