def calc_insert_hist(self):
counter = 0
skip = 0
skip_counter = 0
mads = 10
ins_list = []
# Each entry in valueCounts is a value, and its count is
# the number of instances of that value observed in the dataset.
# So valueCount[5] is the number of times 5 has been seen in the data.
valueCounts = Counter()
for read in self.bam.fetch():
if skip_counter < skip:
skip_counter += 1
continue
if (read.is_reverse
or not read.mate_is_reverse
or read.is_unmapped
or read.mate_is_unmapped
or not self.is_primary(read)
or read.template_length <= 0
or read.get_tag('RG') not in self.readgroups):
continue
else:
valueCounts[read.template_length] += 1
counter += 1
if counter == self.num_samp:
break
if len(valueCounts) == 0:
sys.stderr.write('Error: failed to build insert size histogram for paired-end reads.\n\
Please ensure BAM file (%s) has inward facing, paired-end reads.\n' % self.bam.filename)
exit(1)
# remove outliers
med = median(valueCounts)
u_mad = upper_mad(valueCounts, med)
for x in [x for x in list(valueCounts) if x > med + mads * u_mad]:
del valueCounts[x]
self.hist = valueCounts
self.mean = mean(self.hist)
self.sd = stdev(self.hist)
def calc_insert_hist(self):
counter = 0
skip = 0
skip_counter = 0
mads = 10
ins_list = []
# Each entry in valueCounts is a value, and its count is
# the number of instances of that value observed in the dataset.
# So valueCount[5] is the number of times 5 has been seen in the data.
valueCounts = Counter()
for read in self.bam.fetch():
if skip_counter < skip:
skip_counter += 1
continue
if (read.is_reverse
or not read.mate_is_reverse
or read.is_unmapped
or read.mate_is_unmapped
or not self.is_primary(read)
or read.template_length <= 0
or read.get_tag('RG') not in self.readgroups):
continue
else:
valueCounts[read.template_length] += 1
counter += 1
if counter == self.num_samp:
break
if len(valueCounts) == 0:
sys.stderr.write('Error: failed to build insert size histogram for paired-end reads.\n\
Please ensure BAM file (%s) has inward facing, paired-end reads.\n' % self.bam.filename)
exit(1)
# remove outliers
med = median(valueCounts)
u_mad = upper_mad(valueCounts, med)
for x in [x for x in list(valueCounts) if x > med + mads * u_mad]:
del valueCounts[x]
self.hist = valueCounts
self.mean = mean(self.hist)
self.sd = stdev(self.hist)