def check_dna_chars_primers(header,
mapping_data,
errors,
disable_primer_check=False
):
""" Checks for valid DNA characters in primer fields
Also flags empty fields as errors unless flags are passed to suppress
barcode or primer checks.
header: list of header strings
mapping_data: list of lists of raw metadata mapping file data
errors: list of errors
disable_primer_check: If True, disables tests for valid primer sequences.
"""
valid_dna_chars = DNASequence.iupac_characters()
valid_dna_chars.add(',')
# Detect fields directly, in case user does not have fields in proper
# order in the mapping file (this will generate error separately)
header_fields_to_check = ["ReversePrimer"]
if not disable_primer_check:
header_fields_to_check.append("LinkerPrimerSequence")
check_indices = []
for curr_field in range(len(header)):
if header[curr_field] in header_fields_to_check:
check_indices.append(curr_field)
# Correction factor for header being the first line
correction_ix = 1
# Check for missing data
for curr_data in range(len(mapping_data)):
for curr_ix in check_indices:
if len(mapping_data[curr_data][curr_ix]) == 0:
errors.append("Missing expected DNA sequence\t%d,%d" %
(curr_data + correction_ix, curr_ix))
# Check for non-DNA characters
for curr_data in range(len(mapping_data)):
for curr_ix in check_indices:
for curr_nt in mapping_data[curr_data][curr_ix]:
if curr_nt not in valid_dna_chars:
errors.append("Invalid DNA sequence detected: %s\t%d,%d" %
(mapping_data[curr_data][curr_ix],
curr_data + correction_ix, curr_ix))
continue
return errors
def check_dna_chars_primers(header,
mapping_data,
errors,
disable_primer_check=False):
""" Checks for valid DNA characters in primer fields
Also flags empty fields as errors unless flags are passed to suppress
barcode or primer checks.
header: list of header strings
mapping_data: list of lists of raw metadata mapping file data
errors: list of errors
disable_primer_check: If True, disables tests for valid primer sequences.
"""
valid_dna_chars = DNASequence.iupac_characters()
valid_dna_chars.add(',')
# Detect fields directly, in case user does not have fields in proper
# order in the mapping file (this will generate error separately)
header_fields_to_check = ["ReversePrimer"]
if not disable_primer_check:
header_fields_to_check.append("LinkerPrimerSequence")
check_indices = []
for curr_field in range(len(header)):
if header[curr_field] in header_fields_to_check:
check_indices.append(curr_field)
# Correction factor for header being the first line
correction_ix = 1
# Check for missing data
for curr_data in range(len(mapping_data)):
for curr_ix in check_indices:
if len(mapping_data[curr_data][curr_ix]) == 0:
errors.append("Missing expected DNA sequence\t%d,%d" %
(curr_data + correction_ix, curr_ix))
# Check for non-DNA characters
for curr_data in range(len(mapping_data)):
for curr_ix in check_indices:
for curr_nt in mapping_data[curr_data][curr_ix]:
if curr_nt not in valid_dna_chars:
errors.append("Invalid DNA sequence detected: %s\t%d,%d" %
(mapping_data[curr_data][curr_ix],
curr_data + correction_ix, curr_ix))
continue
return errors
def get_consensus(fasta_tempfile, min_consensus):
"""
Returns consensus sequence from a set of sequences
input: fasta file, min_consensus
fasta_file should be in the following format:
>random_bc|number
seq
>random_bc|number
seq
....
number = number of times this seq has appeared with this random_barcode
Parameters
----------
fasta_seqs: list
min_consensus: float
Returns
----------
consensus_seq: string
consensus sequence for the given list of sequences
"""
seqs = list()
counts = list()
for label, seq in parse_fasta(fasta_tempfile):
RE_output = search(r'\w+\|(\d+)', label)
counts.append(int(RE_output.group(1)))
seqs.append(seq)
length = len(seqs[0])
number_of_seqs = len(seqs)
for seq_index in range(number_of_seqs):
if len(seqs[seq_index]) != length:
raise SeqLengthMismatchError()
freq_this_pos_this_base = dict()
count_of_seq_with_max_count = dict()
for x in range(length):
freq_this_pos_this_base[x] = dict()
count_of_seq_with_max_count[x] = dict()
for y in DNASequence.iupac_characters():
freq_this_pos_this_base[x][y] = 0
count_of_seq_with_max_count[x][y] = 0
for this_seq_count, seq in enumerate(seqs):
freq_this_pos_this_base[x][
seq[x]] += counts[this_seq_count]
if counts[this_seq_count] > count_of_seq_with_max_count[x][seq[x]]:
count_of_seq_with_max_count[x][seq[x]] = counts[this_seq_count]
consensus = list()
for index in range(length):
sorted_bases = sorted(
freq_this_pos_this_base[index].iteritems(),
key=lambda x: x[1])
max_base, max_freq = sorted_bases[-1]
for (counter, (b, n)) in enumerate(sorted_bases):
if max_freq == n:
try:
if (count_of_seq_with_max_count[counter][b] >
count_of_seq_with_max_count[counter][max_base]):
max_base = b
except KeyError:
pass
score = 10.0 * max_freq / number_of_seqs
if score < min_consensus:
raise LowConsensusScoreError()
consensus.append(max_base)
consensus_seq = ''.join(map(str, consensus))
return consensus_seq
