def _process_read(self, read):
while len(self.stats["positions"]) < len(read[1]):
self.stats["positions"].append(deepcopy(self.initial_dict))
for pos in range(len(read[1])):
base = read[1][pos]
score = pyngsqc.get_qual_from_phred(read[3][pos], self.qual_offset)
self.stats["positions"][pos]["scores"][score] += 1
try:
self.stats["positions"][pos]["bases"][base] += 1
except KeyError:
self.stats["positions"][pos]["bases"]["N"] += 1
def _process_read(self, read):
while len(self.stats["positions"]) < len(read[1]):
self.stats["positions"].append(deepcopy(self.initial_dict))
for pos in xrange(len(read[1])):
base = read[1][pos]
score = pyngsqc.get_qual_from_phred(read[3][pos], self.qual_offset)
self.stats["positions"][pos]["scores"][score] += 1
try:
self.stats["positions"][pos]["bases"][base] += 1
except KeyError:
self.stats["positions"][pos]["bases"]["N"] += 1
def trim_read(self, read):
for iii in reversed(xrange(len(read[1]))):
if self.remove_trailing_Ns and read[1][iii].upper() == "N":
# Remove last base and score
read[1] = read[1][:-1] # Base
read[3] = read[3][:-1] # Phred Score
elif pyngsqc.get_qual_from_phred(read[3][iii], self.qual_offset) \
< self.qual_threshold:
read[1] = read[1][:-1] # Base
read[3] = read[3][:-1] # Phred Score
if len(read[1]) < self.min_length:
return ()
else:
return read
def _passes_qc(read):
qual = read[3]
read_len = len(qual)
low_scores = 0
for p in qual:
if pyngsqc.get_qual_from_phred(p, self.qual_offset) < self.qual_threshold:
low_scores += 1
this_pass_rate = 1.0 - float(low_scores) / float(read_len)
if this_pass_rate <= self.pass_rate:
return False
elif self.max_Ns != -1 and int(pyngsqc.num_Ns_in_read(read)) > self.max_Ns:
return False
else:
return True
def _passes_qc(read):
qual = read[3]
read_len = len(qual)
low_scores = 0
for p in qual:
if pyngsqc.get_qual_from_phred(p, self.qual_offset) <\
self.qual_threshold:
low_scores += 1
this_pass_rate = 1.0 - float(low_scores) / float(read_len)
if this_pass_rate <= self.pass_rate:
return False
elif self.max_Ns != -1 and \
int(pyngsqc.num_Ns_in_read(read)) > self.max_Ns:
return False
else:
return True