def test_get_LEA_seq_consensus_seqs(self):
barcode_type = int(7)
barcode_len = 7
barcode_correction_fn = None
max_barcode_errors = 1.5
min_consensus = 0.66
max_cluster_ratio = 2.5
min_difference_in_bcs = 0.86
fwd_length = 19
rev_length = 19
min_reads_per_random_bc = 1
min_diff_in_clusters = self.min_difference_in_clusters
barcode_column = 'BarcodeSequence'
reverse_primer_column = 'ReversePrimer'
function_call, _ = get_LEA_seq_consensus_seqs(
self.fwd_read_fh, self.rev_read_fh, self.mapping_fp, self.temp_dir,
barcode_type, barcode_len, barcode_correction_fn,
max_barcode_errors, min_consensus, max_cluster_ratio,
min_difference_in_bcs, fwd_length, rev_length,
min_reads_per_random_bc, min_diff_in_clusters, barcode_column,
reverse_primer_column)
actual = function_call['Sample1']['AGCTACGAGCTATTGC']
expected = 'AAAAAAAAAAAAAAAAAAA^AAAAAAAAAAAAAAAAAA'
self.assertEqual(actual, expected)
# this call tests the second condition of if loop
# in the function get_consensus_seq_lookup
# i.e. select the majority sequence, as the cluster ratio
# between max_cluster/second_best_cluster in the fwd_read_data
# (and rev_read_data) is 3/1 > 2.5,
# so the function get_consensus will not be called
fn_call, _ = get_LEA_seq_consensus_seqs(
self.get_cons_fwd_read_fh, self.get_cons_rev_read_fh,
self.get_cons_mapping_fp, self.temp_dir, barcode_type, barcode_len,
barcode_correction_fn, max_barcode_errors, min_consensus,
max_cluster_ratio, min_difference_in_bcs, fwd_length, rev_length,
min_reads_per_random_bc, min_diff_in_clusters, barcode_column,
reverse_primer_column)
get_cons_actual = fn_call['Sample1']['AGCTACGAGCTATTGC']
get_cons_expected = 'AAAAAAAAAACAAAAAAAA^AAAAAAAAAATAAAAATA'
self.assertEqual(get_cons_actual, get_cons_expected)
# this call tests the third condition of if loop
# in the function get_consensus_seq_lookup.
# i.e. calls the get_consensus function, as the cluster ratio
# between max_cluster/second_best_cluster in the get_cons_fwd_read_data
# (and get_cons_rev_read_data) is 2/1 ( < 2.5)
# so the majority sequence will not be selected
get_cons_actual = fn_call['Sample2']['AGCTACGCATCAAGGG']
get_cons_expected = 'AAAAAAAAAATAAAAAAAA^TTAAAAAAAAAAAAGAAAA'
self.assertEqual(get_cons_actual, get_cons_expected)
self.assertFalse(len(fn_call) <= 1,
msg="The get_consensus_seqs_lookup function "
"has returned early, without completing "
"the three 'for' loops.")
def test_get_LEA_seq_consensus_seqs(self):
barcode_type = int(7)
barcode_len = 7
barcode_correction_fn = None
max_barcode_errors = 1.5
min_consensus = 0.66
max_cluster_ratio = 2.5
min_difference_in_bcs = 0.86
fwd_length = 19
rev_length = 19
min_reads_per_random_bc = 1
min_diff_in_clusters = self.min_difference_in_clusters
barcode_column = 'BarcodeSequence'
reverse_primer_column = 'ReversePrimer'
function_call, _ = get_LEA_seq_consensus_seqs(self.fwd_read_data,
self.rev_read_data,
self.mapping_fp,
self.temp_dir,
barcode_type,
barcode_len,
barcode_correction_fn,
max_barcode_errors,
min_consensus,
max_cluster_ratio,
min_difference_in_bcs,
fwd_length,
rev_length,
min_reads_per_random_bc,
min_diff_in_clusters,
barcode_column,
reverse_primer_column)
actual = function_call['Sample1']['AGCTACGAGCTATTGC']
expected = 'AAAAAAAAAAAAAAAAAAA^AAAAAAAAAAAAAAAAAA'
self.assertEqual(actual, expected)
def test_get_LEA_seq_consensus_seqs(self):
barcode_type = int(7)
barcode_len = 7
barcode_correction_fn = None
max_barcode_errors = 1.5
min_consensus = 0.66
max_cluster_ratio = 2.5
min_difference_in_bcs = 0.86
fwd_length = 19
rev_length = 19
min_reads_per_random_bc = 1
min_diff_in_clusters = self.min_difference_in_clusters
barcode_column = 'BarcodeSequence'
reverse_primer_column = 'ReversePrimer'
function_call, _ = get_LEA_seq_consensus_seqs(self.fwd_read_fh,
self.rev_read_fh,
self.mapping_fp,
self.temp_dir,
barcode_type,
barcode_len,
barcode_correction_fn,
max_barcode_errors,
min_consensus,
max_cluster_ratio,
min_difference_in_bcs,
fwd_length,
rev_length,
min_reads_per_random_bc,
min_diff_in_clusters,
barcode_column,
reverse_primer_column)
actual = function_call['Sample1']['AGCTACGAGCTATTGC']
expected = 'AAAAAAAAAAAAAAAAAAA^AAAAAAAAAAAAAAAAAA'
self.assertEqual(actual, expected)
# this call tests the second condition of if loop
# in the function get_consensus_seq_lookup
# i.e. select the majority sequence, as the cluster ratio
# between max_cluster/second_best_cluster in the fwd_read_data
# (and rev_read_data) is 3/1 > 2.5,
# so the function get_consensus will not be called
fn_call, _ = get_LEA_seq_consensus_seqs(self.get_cons_fwd_read_fh,
self.get_cons_rev_read_fh,
self.get_cons_mapping_fp,
self.temp_dir,
barcode_type,
barcode_len,
barcode_correction_fn,
max_barcode_errors,
min_consensus,
max_cluster_ratio,
min_difference_in_bcs,
fwd_length,
rev_length,
min_reads_per_random_bc,
min_diff_in_clusters,
barcode_column,
reverse_primer_column)
get_cons_actual = fn_call['Sample1']['AGCTACGAGCTATTGC']
get_cons_expected = 'AAAAAAAAAACAAAAAAAA^AAAAAAAAAATAAAAATA'
self.assertEqual(get_cons_actual, get_cons_expected)
# this call tests the third condition of if loop
# in the function get_consensus_seq_lookup.
# i.e. calls the get_consensus function, as the cluster ratio
# between max_cluster/second_best_cluster in the get_cons_fwd_read_data
# (and get_cons_rev_read_data) is 2/1 ( < 2.5)
# so the majority sequence will not be selected
get_cons_actual = fn_call['Sample2']['AGCTACGCATCAAGGG']
get_cons_expected = 'AAAAAAAAAATAAAAAAAA^TTAAAAAAAAAAAAGAAAA'
self.assertEqual(get_cons_actual, get_cons_expected)
self.assertFalse(len(fn_call) <= 1,
msg="The get_consensus_seqs_lookup function "
"has returned early, without completing "
"the three 'for' loops.")
def main():
option_parser, opts, args = parse_command_line_parameters(**script_info)
barcode_type = opts.barcode_type
max_barcode_errors = opts.max_barcode_errors
mapping_fp = opts.mapping_fp
sequence_read_fps = opts.sequence_read_fps
min_consensus = opts.min_consensus
max_cluster_ratio = opts.max_cluster_ratio
output_dir = opts.output_dir
min_difference_in_bcs = opts.min_difference_in_bcs
fwd_length = opts.fwd_length
rev_length = opts.rev_length
min_reads_per_random_bc = opts.min_reads_per_random_bc
min_diff_in_clusters = opts.min_difference_in_clusters
barcode_column = opts.header_barcode_column
reverse_primer_column = opts.reverse_primer_column
create_dir(output_dir)
fwd_consensus_outfile = open(path.join(output_dir, "fwd.fna"), "w")
rev_consensus_outfile = open(path.join(output_dir, "rev.fna"), "w")
log_file = open(path.join(output_dir, "log.txt"), "w")
if barcode_type == 'golay_12':
barcode_correction_fn = decode_golay_12
barcode_len = 12
else:
barcode_correction_fn = None
try:
barcode_len = int(barcode_type)
except ValueError:
option_parser.error("Invalid barcode type '%s'. The barcode type "
"must be either golay_12 or a positive "
"integer indicating the barcode length." %
barcode_type)
if max_barcode_errors < 0:
option_parser.error("--max_barcode_errors must be greater than or "
"equal to zero. You provided %.4f." %
max_barcode_errors)
if barcode_len < 1:
option_parser.error("Invalid barcode length: %d. Must be greater "
"than zero." % barcode_len)
if len(sequence_read_fps) != 2:
option_parser.error("You must provide exactly two sequence read "
"filepaths, the first for forward reads and "
"second for reverse reads. You specified %d "
"filepaths." % len(sequence_read_fps))
fwd_read_f = open(sequence_read_fps[0], 'U')
rev_read_f = open(sequence_read_fps[1], 'U')
map_f = open(mapping_fp, 'U')
(consensus_seq_lookup,
log_out) = get_LEA_seq_consensus_seqs(fwd_read_f,
rev_read_f,
map_f,
output_dir,
barcode_type,
barcode_len,
barcode_correction_fn,
max_barcode_errors,
min_consensus,
max_cluster_ratio,
min_difference_in_bcs,
fwd_length,
rev_length,
min_reads_per_random_bc,
min_diff_in_clusters,
barcode_column,
reverse_primer_column)
for sample_id in consensus_seq_lookup:
for bc_index, rand_bc in enumerate(consensus_seq_lookup[sample_id]):
consensus_seq = consensus_seq_lookup[sample_id][rand_bc]
fwd_consensus, rev_consensus = consensus_seq.split('^')
fwd_consensus_outfile.write(">{}_{}\n{}\n".format(
sample_id, bc_index, fwd_consensus))
rev_consensus_outfile.write(">{}_{}\n{}\n".format(
sample_id, bc_index, rev_consensus))
log_file.write(log_out)
log_file.close()
fwd_read_f.close()
rev_read_f.close()
fwd_consensus_outfile.close()
rev_consensus_outfile.close()
map_f.close()
def main():
option_parser, opts, args = parse_command_line_parameters(**script_info)
barcode_type = opts.barcode_type
max_barcode_errors = opts.max_barcode_errors
mapping_fp = opts.mapping_fp
sequence_read_fps = opts.sequence_read_fps
min_consensus = opts.min_consensus
max_cluster_ratio = opts.max_cluster_ratio
output_dir = opts.output_dir
min_difference_in_bcs = opts.min_difference_in_bcs
fwd_length = opts.fwd_length
rev_length = opts.rev_length
min_reads_per_random_bc = opts.min_reads_per_random_bc
min_diff_in_clusters = opts.min_difference_in_clusters
barcode_column = opts.header_barcode_column
reverse_primer_column = opts.reverse_primer_column
create_dir(output_dir)
fwd_consensus_outfile = open(path.join(output_dir, "fwd.fna"), "w")
rev_consensus_outfile = open(path.join(output_dir, "rev.fna"), "w")
log_file = open(path.join(output_dir, "log.txt"), "w")
if barcode_type == 'golay_12':
barcode_correction_fn = decode_golay_12
barcode_len = 12
else:
barcode_correction_fn = None
try:
barcode_len = int(barcode_type)
except ValueError:
option_parser.error("Invalid barcode type '%s'. The barcode type "
"must be either golay_12 or a positive "
"integer indicating the barcode length." %
barcode_type)
if max_barcode_errors < 0:
option_parser.error("--max_barcode_errors must be greater than or "
"equal to zero. You provided %.4f." %
max_barcode_errors)
if min_diff_in_clusters < 0 or min_diff_in_clusters > 1:
option_parser.error("--min_difference_in_clusters must be "
"between 0 to 1. You provided %.4f." %
min_diff_in_clusters)
if min_difference_in_bcs < 0 or min_difference_in_bcs > 1:
option_parser.error("--min_difference_in_bcs must be between 0 to 1."
" You provided %.4f." % min_difference_in_bcs)
if barcode_len < 1:
option_parser.error("Invalid barcode length: %d. Must be greater "
"than zero." % barcode_len)
if len(sequence_read_fps) != 2:
option_parser.error("You must provide exactly two sequence read "
"filepaths, the first for forward reads and "
"second for reverse reads. You specified %d "
"filepaths." % len(sequence_read_fps))
fwd_read_f = open(sequence_read_fps[0], 'U')
rev_read_f = open(sequence_read_fps[1], 'U')
map_f = open(mapping_fp, 'U')
(consensus_seq_lookup, log_out) = get_LEA_seq_consensus_seqs(
fwd_read_f, rev_read_f, map_f, output_dir, barcode_type, barcode_len,
barcode_correction_fn, max_barcode_errors, min_consensus,
max_cluster_ratio, min_difference_in_bcs, fwd_length, rev_length,
min_reads_per_random_bc, min_diff_in_clusters, barcode_column,
reverse_primer_column)
for sample_id in consensus_seq_lookup:
for bc_index, rand_bc in enumerate(consensus_seq_lookup[sample_id]):
consensus_seq = consensus_seq_lookup[sample_id][rand_bc]
fwd_consensus, rev_consensus = consensus_seq.split('^')
fwd_consensus_outfile.write(">{}_{}\n{}\n".format(
sample_id, bc_index, fwd_consensus))
rev_consensus_outfile.write(">{}_{}\n{}\n".format(
sample_id, bc_index, rev_consensus))
log_file.write(log_out)
log_file.close()
fwd_read_f.close()
rev_read_f.close()
fwd_consensus_outfile.close()
rev_consensus_outfile.close()
map_f.close()
