def main():
args = get_args()
# setup logging
log, my_name = setup_logging(args.verbosity, args.log_path)
text = " Starting {} ".format(my_name)
log.info(text.center(65, "="))
alignments = []
log.info("Getting aligned sequences for trimming")
for ftype in get_file_extensions(args.input_format):
alignments.extend(glob.glob(os.path.join(args.input, "*{}".format(ftype))))
# package up needed arguments for map()
package = [args.input_format, args.window, args.threshold, args.proportion, args.max_divergence, args.min_length]
params = zip([package] * len(alignments), alignments)
log.info("Alignment begins. 'X' indicates dropped alignments (these are reported after alignment)")
# if --multprocessing, use Pool.map(), else use map()
# can also extend to MPI map, but not really needed on multicore
# machine
if args.cores > 1:
assert args.cores <= multiprocessing.cpu_count(), "You've specified more cores than you have"
pool = multiprocessing.Pool(args.cores - 1)
alignments = pool.map(get_and_trim_alignments, params)
else:
alignments = map(get_and_trim_alignments, params)
# kick the stdout down one line since we were using sys.stdout
print("")
# drop back into logging
log.info("Alignment ends")
# write the output files
write_alignments_to_outdir(log, args.output, alignments, args.output_format)
# end
text = " Completed {} ".format(my_name)
log.info(text.center(65, "="))
def main(args):
# setup logging
log, my_name = setup_logging(args)
# create the fasta dictionary
loci = get_fasta_dict(log, args)
log.info("Aligning with {}".format(str(args.aligner).upper()))
opts = [[args.window, args.threshold, args.no_trim, args.proportion, args.max_divergence, args.min_length] \
for i in range(len(loci))]
# combine loci and options
params = zip(loci.items(), opts)
log.info("Alignment begins. 'X' indicates dropped alignments (these are reported after alignment)")
# During alignment, drop into sys.stdout for progress indicator
# because logging in multiprocessing is more painful than what
# we really need. Return to logging when alignment completes.
if args.cores > 1:
assert args.cores <= multiprocessing.cpu_count(), "You've specified more cores than you have"
pool = multiprocessing.Pool(args.cores)
alignments = pool.map(align, params)
else:
alignments = map(align, params)
# kick the stdout down one line since we were using sys.stdout
print("")
# drop back into logging
log.info("Alignment ends")
# write the output files
write_alignments_to_outdir(log, args.output, alignments, args.output_format)
# end
text = " Completed {} ".format(my_name)
log.info(text.center(65, "="))
def main(args):
if args.aligner == "muscle":
from phyluce.muscle import Align as align_class
elif args.aligner == "mafft":
from phyluce.mafft import Align as align_class
# create the fasta dictionary
loci = get_fasta_dict(log, args)
log.info("Aligning with {}".format(str(args.aligner).upper()))
opts = [[args.window, args.threshold, args.no_trim, args.proportion, args.max_divergence, args.min_length, align_class] \
for _ in loci]
# combine loci and options
params = zip(loci.items(), opts)
log.info("Alignment begins. 'X' indicates dropped alignments (these are reported after alignment)")
# During alignment, drop into sys.stdout for progress indicator
# because logging in multiprocessing is more painful than what
# we really need. Return to logging when alignment completes.
if args.cores > 1:
assert args.cores <= multiprocessing.cpu_count(), "You've specified more cores than you have"
pool = multiprocessing.Pool(args.cores)
alignments = pool.map(align, params)
else:
alignments = map(align, params)
#import pickle
#with open('/Users/tobias/Desktop/alignments.pickle', 'wb') as handle:
# pickle.dump(alignments, handle, protocol=pickle.HIGHEST_PROTOCOL)
#with open('/Users/tobias/Desktop/alignments.pickle', 'rb') as handle:
# alignments = pickle.load(handle)
# kick the stdout down one line since we were using sys.stdout
print("")
# drop back into logging
log.info("Alignment ends")
# write the output files
for name, alignment in alignments:
if alignment.trimmed:
for t in alignment.trimmed:
t.id = t.id.split('_', 3)[3]
t.name = t.id
t.description = ''
write_alignments_to_outdir(log, args.output, alignments, args.output_format)
# end
text = " Completed! "
log.info(text.center(65, "="))
def main(args):
if args.aligner == "muscle":
from phyluce.muscle import Align as align_class
elif args.aligner == "mafft":
from phyluce.mafft import Align as align_class
# create the fasta dictionary
loci = get_fasta_dict(log, args)
log.info("Aligning with {}".format(str(args.aligner).upper()))
opts = [[
args.window, args.threshold, args.no_trim, args.proportion,
args.max_divergence, args.min_length, args.gap_opening_penalty,
args.gap_extension_penalty, align_class
] for _ in loci]
# combine loci and options
params = zip(loci.items(), opts)
log.info(
"Alignment begins. 'X' indicates dropped alignments (these are reported after alignment)"
)
# During alignment, drop into sys.stdout for progress indicator
# because logging in multiprocessing is more painful than what
# we really need. Return to logging when alignment completes.
if args.cores > 1:
assert args.cores <= multiprocessing.cpu_count(
), "You've specified more cores than you have"
pool = multiprocessing.Pool(args.cores)
alignments = pool.map(align, params)
else:
alignments = map(align, params)
#import pickle
#with open('/Users/tobias/Desktop/alignments.pickle', 'wb') as handle:
# pickle.dump(alignments, handle, protocol=pickle.HIGHEST_PROTOCOL)
#with open('/Users/tobias/Desktop/alignments.pickle', 'rb') as handle:
# alignments = pickle.load(handle)
# kick the stdout down one line since we were using sys.stdout
print("")
# drop back into logging
log.info("Alignment ends")
# write the output files
for name, alignment in alignments:
if alignment.trimmed:
for t in alignment.trimmed:
locus_name = t.description.split('|')[-1]
rest_of_string = '|'.join(t.description.split('|')[:-1])
string_to_replace = '%s_' % str(locus_name)
new_string = t.id
# fix the fasta header, also removing the occasional _R_ resulting from reverse contigs
tmp = re.sub(string_to_replace, '', new_string, 1)
t.id = re.sub('_R_', '', tmp, 1)
t.name = t.id
t.description = ''
write_alignments_to_outdir(log, args.output, alignments,
args.output_format)
try:
#input_folder = '/'.join(args.sequences.split('/')[:-2])
pickle_path = os.path.join(args.output, '.secapr_files')
if not os.path.exists(pickle_path):
os.makedirs(pickle_path)
with open(os.path.join(pickle_path, 'sequence_origin.pickle'),
'wb') as handle:
pickle.dump(args.sequences,
handle,
protocol=pickle.HIGHEST_PROTOCOL)
# end
text = " Completed! "
log.info(text.center(65, "="))
except:
print('Could not pass origin of sequences to %s' % pickle_path)