def test_safe_md5(self):
"""Make sure we have the expected md5"""
exp = 'ab07acbb1e496801937adfa772424bf7'
fd = BytesIO(b'foo bar baz')
obs = safe_md5(fd)
self.assertEqual(obs.hexdigest(), exp)
fd.close()
def test_safe_md5(self):
"""Make sure we have the expected md5"""
exp = 'ab07acbb1e496801937adfa772424bf7'
fd = BytesIO(b'foo bar baz')
obs = safe_md5(fd)
self.assertEqual(obs.hexdigest(), exp)
fd.close()
def generate_run_xml(self):
"""Generates the run XML file
Returns
-------
xml.etree.Element
The root elelement of the generated ``ElementTree``
"""
run_set = ET.Element('RUN_SET', {
"xmlns:xsi": "http://www.w3.org/2001/XMLSchema-instance",
"xsi:noNamespaceSchemaLocation": "ftp://ftp.sra.ebi.ac.uk/meta/xsd"
"/sra_1_3/SRA.run.xsd"})
for sample_name, sample_info in sorted(viewitems(self.samples)):
for row_number, prep_info in enumerate(sample_info['preps']):
experiment_alias = self._get_experiment_alias(sample_name,
row_number)
file_type = prep_info['file_type']
file_path = prep_info['file_path']
with open(file_path) as fp:
md5 = safe_md5(fp).hexdigest()
run = ET.SubElement(run_set, 'RUN', {
'alias': basename(file_path) + '_run',
'center_name': 'CCME-COLORADO'}
)
ET.SubElement(run, 'EXPERIMENT_REF', {
'refname': experiment_alias}
)
data_block = ET.SubElement(run, 'DATA_BLOCK')
files = ET.SubElement(data_block, 'FILES')
ET.SubElement(files, 'FILE', {
'filename': basename(file_path),
'filetype': file_type,
'quality_scoring_system': 'phred',
'checksum_method': 'MD5',
'checksum': md5}
)
return run_set
def main():
option_parser, opts, args = parse_command_line_parameters(**script_info)
sequence_read_fps = opts.sequence_read_fps
barcode_read_fps = opts.barcode_read_fps
sample_ids = None
if opts.sample_ids is not None:
sample_ids = opts.sample_ids.split(',')
mapping_fps = opts.mapping_fps
phred_quality_threshold = opts.phred_quality_threshold
retain_unassigned_reads = opts.retain_unassigned_reads
min_per_read_length_fraction = opts.min_per_read_length_fraction
max_bad_run_length = opts.max_bad_run_length
rev_comp = opts.rev_comp
rev_comp_barcode = opts.rev_comp_barcode
rev_comp_mapping_barcodes = opts.rev_comp_mapping_barcodes
seq_max_N = opts.sequence_max_n
start_seq_id = opts.start_seq_id
# NEED TO FIX THIS FUNCTIONALITY - CURRENTLY READING THE WRONG FIELD
# opts.filter_bad_illumina_qual_digit
filter_bad_illumina_qual_digit = False
store_qual_scores = opts.store_qual_scores
store_demultiplexed_fastq = opts.store_demultiplexed_fastq
barcode_type = opts.barcode_type
max_barcode_errors = opts.max_barcode_errors
# if this is not a demultiplexed run,
if barcode_type == 'not-barcoded':
if sample_ids is None:
option_parser.error("If not providing barcode reads (because "
"your data is not multiplexed), must provide --sample_ids.")
if len(sample_ids) != len(sequence_read_fps):
option_parser.error("If providing --sample_ids (because "
"your data is not multiplexed), must provide the same number "
"of sample ids as sequence read filepaths.")
barcode_read_fps = [None] * len(sequence_read_fps)
mapping_fps = [None] * len(sequence_read_fps)
elif barcode_read_fps is None:
option_parser.error("Must provide --barcode_read_fps if "
"--barcode_type is not 'not-barcoded'")
elif mapping_fps is None:
option_parser.error("Must provide --mapping_fps if "
"--barcode_type is not 'not-barcoded'")
phred_offset = opts.phred_offset
if phred_offset is not None:
try:
phred_to_ascii_f = phred_to_ascii_fs[phred_offset]
except KeyError:
# shouldn't be able to get here, but we'll stay on the
# safe side
option_parser.error("Only valid phred offsets are: %s" %
' '.join(phred_to_ascii_fs.keys()))
else:
# let split_libraries_fastq.process_fastq_single_end_read_file
# figure it out...
phred_to_ascii_f = None
if opts.last_bad_quality_char is not None:
option_parser.error('--last_bad_quality_char is no longer supported. '
'Use -q instead (see option help text by passing -h)')
if not (0 <= min_per_read_length_fraction <= 1):
option_parser.error('--min_per_read_length_fraction must be between '
'0 and 1 (inclusive). You passed %1.5f' %
min_per_read_length_fraction)
barcode_correction_fn = BARCODE_DECODER_LOOKUP.get(barcode_type, None)
if len(mapping_fps) == 1 and len(sequence_read_fps) > 1:
mapping_fps = mapping_fps * len(sequence_read_fps)
if len(set([len(sequence_read_fps), len(barcode_read_fps),
len(mapping_fps)])) > 1:
option_parser.error("Same number of sequence, barcode, and mapping "
"files must be provided.")
output_dir = opts.output_dir
create_dir(output_dir)
output_fp_temp = '%s/seqs.fna.incomplete' % output_dir
output_fp = '%s/seqs.fna' % output_dir
output_f = open(output_fp_temp, 'w')
qual_fp_temp = '%s/qual.fna.incomplete' % output_dir
qual_fp = '%s/seqs.qual' % output_dir
output_fastq_fp_temp = '%s/seqs.fastq.incomplete' % output_dir
output_fastq_fp = '%s/seqs.fastq' % output_dir
if store_qual_scores:
qual_f = open(qual_fp_temp, 'w')
# define a qual writer whether we're storing
# qual strings or not so we don't have to check
# every time through the for loop below
def qual_writer(h, q):
qual_f.write('>%s\n%s\n' % (h, q))
else:
def qual_writer(h, q):
pass
if store_demultiplexed_fastq:
output_fastq_f = open(output_fastq_fp_temp, 'w')
# define a fastq writer whether we're storing
# qual strings or not so we don't have to check
# every time through the for loop below
def fastq_writer(h, s, q):
output_fastq_f.write('@%s\n%s\n+\n%s\n' % (h, s, q))
else:
def fastq_writer(h, s, q):
pass
log_fp = '%s/split_library_log.txt' % output_dir
log_f = open(log_fp, 'w')
histogram_fp = '%s/histograms.txt' % output_dir
histogram_f = open(histogram_fp, 'w')
for i in range(len(sequence_read_fps)):
sequence_read_fp = sequence_read_fps[i]
barcode_read_fp = barcode_read_fps[i]
mapping_fp = mapping_fps[i]
if mapping_fp is not None:
mapping_f = open(mapping_fp, 'U')
_, _, barcode_to_sample_id, _, _, _, _ = check_map(mapping_f,
disable_primer_check=True,
has_barcodes=barcode_read_fp is not None)
else:
mapping_f = None
barcode_to_sample_id = {}
if rev_comp_mapping_barcodes:
barcode_to_sample_id = {str(DNA(k).rc()): v for k, v in
barcode_to_sample_id.iteritems()}
if barcode_type == 'golay_12':
invalid_golay_barcodes = get_invalid_golay_barcodes(
barcode_to_sample_id.keys())
if len(invalid_golay_barcodes) > 0:
option_parser.error("Some or all barcodes are not valid golay "
"codes. Do they need to be reverse complemented? If these "
"are not golay barcodes pass --barcode_type 12 to disable "
"barcode error correction, or pass --barcode_type # if "
"the barcodes are not 12 base pairs, where # is the size "
"of the barcodes. Invalid codes:\n\t%s" %
' '.join(invalid_golay_barcodes))
log_f.write("Input file paths\n")
if mapping_fp is not None:
log_f.write('Mapping filepath: %s (md5: %s)\n' %
(mapping_fp, safe_md5(open(mapping_fp)).hexdigest()))
log_f.write('Sequence read filepath: %s (md5: %s)\n' %
(sequence_read_fp,
str(safe_md5(open(sequence_read_fp)).hexdigest())))
if sequence_read_fp.endswith('.gz'):
sequence_read_f = gzip_open(sequence_read_fp)
else:
sequence_read_f = open(sequence_read_fp, 'U')
seq_id = start_seq_id
if barcode_read_fp is not None:
log_f.write('Barcode read filepath: %s (md5: %s)\n\n' %
(barcode_read_fp,
safe_md5(open(barcode_read_fp)).hexdigest()))
if barcode_read_fp.endswith('.gz'):
barcode_read_f = gzip_open(barcode_read_fp)
else:
barcode_read_f = open(barcode_read_fp, 'U')
seq_generator = process_fastq_single_end_read_file(
sequence_read_f, barcode_read_f, barcode_to_sample_id,
store_unassigned=retain_unassigned_reads,
max_bad_run_length=max_bad_run_length,
phred_quality_threshold=phred_quality_threshold,
min_per_read_length_fraction=min_per_read_length_fraction,
rev_comp=rev_comp, rev_comp_barcode=rev_comp_barcode,
seq_max_N=seq_max_N, start_seq_id=start_seq_id,
filter_bad_illumina_qual_digit=filter_bad_illumina_qual_digit,
log_f=log_f, histogram_f=histogram_f,
barcode_correction_fn=barcode_correction_fn,
max_barcode_errors=max_barcode_errors,
phred_to_ascii_f=phred_to_ascii_f)
else:
seq_generator = process_fastq_single_end_read_file_no_barcode(
sequence_read_f, sample_ids[i],
store_unassigned=retain_unassigned_reads,
max_bad_run_length=max_bad_run_length,
phred_quality_threshold=phred_quality_threshold,
min_per_read_length_fraction=min_per_read_length_fraction,
rev_comp=rev_comp, seq_max_N=seq_max_N,
start_seq_id=start_seq_id,
filter_bad_illumina_qual_digit=filter_bad_illumina_qual_digit,
log_f=log_f, histogram_f=histogram_f,
phred_to_ascii_f=phred_to_ascii_f)
for fasta_header, sequence, quality, seq_id in seq_generator:
output_f.write('>%s\n%s\n' % (fasta_header, sequence))
qual_writer(fasta_header, quality)
fastq_writer(fasta_header, sequence, quality)
start_seq_id = seq_id + 1
log_f.write('\n---\n\n')
output_f.close()
rename(output_fp_temp, output_fp)
# process the optional output files, as necessary
if store_qual_scores:
qual_f.close()
rename(qual_fp_temp, qual_fp)
if store_demultiplexed_fastq:
output_fastq_f.close()
rename(output_fastq_fp_temp, output_fastq_fp)
def main():
option_parser, opts, args = parse_command_line_parameters(**script_info)
sequence_read_fps = opts.sequence_read_fps
barcode_read_fps = opts.barcode_read_fps
sample_ids = None
if opts.sample_ids is not None:
sample_ids = opts.sample_ids.split(',')
mapping_fps = opts.mapping_fps
phred_quality_threshold = opts.phred_quality_threshold
retain_unassigned_reads = opts.retain_unassigned_reads
min_per_read_length_fraction = opts.min_per_read_length_fraction
max_bad_run_length = opts.max_bad_run_length
rev_comp = opts.rev_comp
rev_comp_barcode = opts.rev_comp_barcode
rev_comp_mapping_barcodes = opts.rev_comp_mapping_barcodes
seq_max_N = opts.sequence_max_n
start_seq_id = opts.start_seq_id
# NEED TO FIX THIS FUNCTIONALITY - CURRENTLY READING THE WRONG FIELD
# opts.filter_bad_illumina_qual_digit
filter_bad_illumina_qual_digit = False
store_qual_scores = opts.store_qual_scores
store_demultiplexed_fastq = opts.store_demultiplexed_fastq
barcode_type = opts.barcode_type
max_barcode_errors = opts.max_barcode_errors
# if this is not a demultiplexed run,
if barcode_type == 'not-barcoded':
if sample_ids is None:
option_parser.error(
"If not providing barcode reads (because "
"your data is not multiplexed), must provide --sample_ids.")
if len(sample_ids) != len(sequence_read_fps):
option_parser.error(
"If providing --sample_ids (because "
"your data is not multiplexed), must provide the same number "
"of sample ids as sequence read filepaths.")
barcode_read_fps = [None] * len(sequence_read_fps)
mapping_fps = [None] * len(sequence_read_fps)
elif barcode_read_fps is None:
option_parser.error("Must provide --barcode_read_fps if "
"--barcode_type is not 'not-barcoded'")
elif mapping_fps is None:
option_parser.error("Must provide --mapping_fps if "
"--barcode_type is not 'not-barcoded'")
phred_offset = opts.phred_offset
if phred_offset is not None:
try:
phred_to_ascii_f = phred_to_ascii_fs[phred_offset]
except KeyError:
# shouldn't be able to get here, but we'll stay on the
# safe side
option_parser.error("Only valid phred offsets are: %s" %
' '.join(phred_to_ascii_fs.keys()))
else:
# let split_libraries_fastq.process_fastq_single_end_read_file
# figure it out...
phred_to_ascii_f = None
if opts.last_bad_quality_char is not None:
option_parser.error(
'--last_bad_quality_char is no longer supported. '
'Use -q instead (see option help text by passing -h)')
if not (0 <= min_per_read_length_fraction <= 1):
option_parser.error('--min_per_read_length_fraction must be between '
'0 and 1 (inclusive). You passed %1.5f' %
min_per_read_length_fraction)
barcode_correction_fn = BARCODE_DECODER_LOOKUP.get(barcode_type, None)
if len(mapping_fps) == 1 and len(sequence_read_fps) > 1:
mapping_fps = mapping_fps * len(sequence_read_fps)
if len(
set([
len(sequence_read_fps),
len(barcode_read_fps),
len(mapping_fps)
])) > 1:
option_parser.error("Same number of sequence, barcode, and mapping "
"files must be provided.")
output_dir = opts.output_dir
create_dir(output_dir)
output_fp_temp = '%s/seqs.fna.incomplete' % output_dir
output_fp = '%s/seqs.fna' % output_dir
output_f = open(output_fp_temp, 'w')
qual_fp_temp = '%s/qual.fna.incomplete' % output_dir
qual_fp = '%s/seqs.qual' % output_dir
output_fastq_fp_temp = '%s/seqs.fastq.incomplete' % output_dir
output_fastq_fp = '%s/seqs.fastq' % output_dir
if store_qual_scores:
qual_f = open(qual_fp_temp, 'w')
# define a qual writer whether we're storing
# qual strings or not so we don't have to check
# every time through the for loop below
def qual_writer(h, q):
qual_f.write('>%s\n%s\n' % (h, q))
else:
def qual_writer(h, q):
pass
if store_demultiplexed_fastq:
output_fastq_f = open(output_fastq_fp_temp, 'w')
# define a fastq writer whether we're storing
# qual strings or not so we don't have to check
# every time through the for loop below
def fastq_writer(h, s, q):
output_fastq_f.write('@%s\n%s\n+\n%s\n' % (h, s, q))
else:
def fastq_writer(h, s, q):
pass
log_fp = '%s/split_library_log.txt' % output_dir
log_f = open(log_fp, 'w')
histogram_fp = '%s/histograms.txt' % output_dir
histogram_f = open(histogram_fp, 'w')
for i in range(len(sequence_read_fps)):
sequence_read_fp = sequence_read_fps[i]
barcode_read_fp = barcode_read_fps[i]
mapping_fp = mapping_fps[i]
if mapping_fp is not None:
mapping_f = open(mapping_fp, 'U')
_, _, barcode_to_sample_id, _, _, _, _ = check_map(
mapping_f,
disable_primer_check=True,
has_barcodes=barcode_read_fp is not None)
else:
mapping_f = None
barcode_to_sample_id = {}
if rev_comp_mapping_barcodes:
barcode_to_sample_id = {
str(DNA(k).rc()): v
for k, v in barcode_to_sample_id.iteritems()
}
if barcode_type == 'golay_12':
invalid_golay_barcodes = get_invalid_golay_barcodes(
barcode_to_sample_id.keys())
if len(invalid_golay_barcodes) > 0:
option_parser.error(
"Some or all barcodes are not valid golay "
"codes. Do they need to be reverse complemented? If these "
"are not golay barcodes pass --barcode_type 12 to disable "
"barcode error correction, or pass --barcode_type # if "
"the barcodes are not 12 base pairs, where # is the size "
"of the barcodes. Invalid codes:\n\t%s" %
' '.join(invalid_golay_barcodes))
log_f.write("Input file paths\n")
if mapping_fp is not None:
log_f.write('Mapping filepath: %s (md5: %s)\n' %
(mapping_fp, safe_md5(open(mapping_fp)).hexdigest()))
log_f.write('Sequence read filepath: %s (md5: %s)\n' %
(sequence_read_fp,
str(safe_md5(open(sequence_read_fp)).hexdigest())))
if sequence_read_fp.endswith('.gz'):
sequence_read_f = gzip_open(sequence_read_fp)
else:
sequence_read_f = open(sequence_read_fp, 'U')
seq_id = start_seq_id
if barcode_read_fp is not None:
log_f.write(
'Barcode read filepath: %s (md5: %s)\n\n' %
(barcode_read_fp, safe_md5(open(barcode_read_fp)).hexdigest()))
if barcode_read_fp.endswith('.gz'):
barcode_read_f = gzip_open(barcode_read_fp)
else:
barcode_read_f = open(barcode_read_fp, 'U')
seq_generator = process_fastq_single_end_read_file(
sequence_read_f,
barcode_read_f,
barcode_to_sample_id,
store_unassigned=retain_unassigned_reads,
max_bad_run_length=max_bad_run_length,
phred_quality_threshold=phred_quality_threshold,
min_per_read_length_fraction=min_per_read_length_fraction,
rev_comp=rev_comp,
rev_comp_barcode=rev_comp_barcode,
seq_max_N=seq_max_N,
start_seq_id=start_seq_id,
filter_bad_illumina_qual_digit=filter_bad_illumina_qual_digit,
log_f=log_f,
histogram_f=histogram_f,
barcode_correction_fn=barcode_correction_fn,
max_barcode_errors=max_barcode_errors,
phred_to_ascii_f=phred_to_ascii_f)
else:
seq_generator = process_fastq_single_end_read_file_no_barcode(
sequence_read_f,
sample_ids[i],
store_unassigned=retain_unassigned_reads,
max_bad_run_length=max_bad_run_length,
phred_quality_threshold=phred_quality_threshold,
min_per_read_length_fraction=min_per_read_length_fraction,
rev_comp=rev_comp,
seq_max_N=seq_max_N,
start_seq_id=start_seq_id,
filter_bad_illumina_qual_digit=filter_bad_illumina_qual_digit,
log_f=log_f,
histogram_f=histogram_f,
phred_to_ascii_f=phred_to_ascii_f)
for fasta_header, sequence, quality, seq_id in seq_generator:
output_f.write('>%s\n%s\n' % (fasta_header, sequence))
qual_writer(fasta_header, quality)
fastq_writer(fasta_header, sequence, quality)
start_seq_id = seq_id + 1
log_f.write('\n---\n\n')
output_f.close()
rename(output_fp_temp, output_fp)
# process the optional output files, as necessary
if store_qual_scores:
qual_f.close()
rename(qual_fp_temp, qual_fp)
if store_demultiplexed_fastq:
output_fastq_f.close()
rename(output_fastq_fp_temp, output_fastq_fp)