from wub.util import seq as seq_util
# Parse command line arguments:
parser = argparse.ArgumentParser(
description='Reverse (but not complement!) sequences and qualities in fastq file.')
parser.add_argument('input_fastq', nargs='?', help='Input fastq (default: stdin).',
type=argparse.FileType('r'), default=sys.stdin)
parser.add_argument('output_fastq', nargs='?', help='Output fastq (default: stdout)',
type=argparse.FileType('w'), default=sys.stdout)
def reverse_seq_records(input_iterator):
"""Reverse SeqRecord objects.
:param input_iterator: Iterator of SeqRecord objects.
:returns: Generator of reversed SeqRecord objects.
:rtype: generator
"""
for record in input_iterator:
yield record[::-1]
if __name__ == '__main__':
args = parser.parse_args()
input_iterator = seq_util.read_seq_records(
args.input_fastq, format='fastq')
output_iterator = reverse_seq_records(input_iterator)
seq_util.write_seq_records(
output_iterator, args.output_fastq, format='fastq')
# Read in chromosomes of the input genome:
chromosomes = list(seq_util.read_seq_records(args.input_fasta))
# Process error weights:
error_weights = np.array(parse_util.separated_list_to_floats(args.w))
# Normalise error weights to probabilities:
error_weights = parse_util.normalise_array(error_weights)
error_weights = dict(
zip(['substitution', 'insertion', 'deletion'], error_weights))
sw = None
if args.s is not None:
sw = sam_writer.SamWriter(args.s, build_sam_header(chromosomes))
simulation_iterator = simulate_sequencing(chromosomes, args.m, args.a,
args.l, args.u, args.e,
error_weights, args.b, args.q,
args.n, sw)
if not args.Q and args.output_fastq != sys.stdout:
simulation_iterator = tqdm.tqdm(simulation_iterator, total=args.n)
for simmed, sam in simulation_iterator:
seq_util.write_seq_records(simmed, args.output_fastq, format='fastq')
if sw is not None:
sw.write(sam)
if sw is not None:
sw.close()
type=argparse.FileType('w'),
default=sys.stdout)
def record_filter(input_iter, in_format, read_names):
""" Filter SeqRecord objects by length and mean quality.
:param input_iter: Iterator of SeqRecord objects.
:param in_format: Input format.
:param to_alphabet: Convert to this alphabet.
:returns: SeqRecord object.
:rtype: generator
"""
for record in input_iter:
if record.id in read_names:
yield record
if __name__ == '__main__':
args = parser.parse_args()
input_iterator = seq_util.read_seq_records(args.input_fastx, format=args.i)
names = args.n.split(',')
output_iterator = record_filter(input_iterator, args.i, names)
seq_util.write_seq_records(output_iterator,
args.output_fastx,
format=args.o)
:param error_weights: Relative frequency of substitutions,insertions,deletions.
:returns: Generator of SeqRecord objects.
:rtype: generator
"""
for record in input_iter:
mutated_seq = sim_seq.simulate_sequencing_errors(record.seq, error_rate, error_weights).seq
record.seq = Seq(mutated_seq)
yield record
if __name__ == '__main__':
args = parser.parse_args()
# Set random seed:
if args.z is not None:
np.random.seed(args.z)
# Process error weights:
error_weights = np.array(parse_util.separated_list_to_floats(args.w))
# Normalise error weights to probabilities:
error_weights = parse_util.normalise_array(error_weights)
error_weights = dict(
zip(['substitution', 'insertion', 'deletion'], error_weights))
input_iterator = seq_util.read_seq_records(args.input_fasta, format='fasta')
simulation_iterator = simulate_errors(input_iterator, args.e, error_weights)
seq_util.write_seq_records(
simulation_iterator, args.output_fasta, format='fasta')
if __name__ == '__main__':
args = parser.parse_args()
input_iterator = seq_util.read_seq_records(args.input_fastx, format=args.i)
if args.b is None:
# Splitting one record per file:
for record in input_iterator:
bn = path.basename(args.input_fastx.name)
ext = bn.rsplit('.', 1)[-1]
fh = open(
path.join(args.output_dir, "{}.{}".format(record.id, ext)),
'w')
seq_util.write_seq_records([record], fh, format=args.o)
fh.flush()
fh.close()
else:
# Split into batches:
input_iterator = batch_iterator(input_iterator, args.b)
i = 0
for records in input_iterator:
bn = path.basename(args.input_fastx.name)
fh = open(path.join(args.output_dir, "batch_{}_{}".format(i, bn)),
'w')
seq_util.write_seq_records(records, fh, format=args.o)
fh.flush()
fh.close()
i += 1