def insert_size(self):
insert = nesteddict()
for sample in self.collect_file:
log = self.collect_file[sample]
line = self.get_line(log, 'MEDIAN_INSERT_SIZE') + 1 # 从1开始
rec = getline(log, line).strip().split()
insert_mean, insert_std = rec[5], rec[6]
insert[sample]['InsertSize_mean'] = insert_mean
insert[sample]['Insert_std'] = insert_std
line = self.get_line(log, 'insert_size')
result = self.read_his(log, line)
histogram = pd.DataFrame.from_dict(result, orient='index').T
histogram['insert_size'] = histogram.index
# histogram=histogram.rename(columns={'count':'Forward_count'})
histogram['count'] = histogram['count'].astype(int)
histogram['insert_size'] = histogram['insert_size'].astype(int)
histogram = histogram.sort_values('insert_size')
histogram['count'] = histogram['count'] / sum(histogram['count'])
histogram.to_csv('{module}/{sample}/insert_size.csv'.format(
module=self.module, sample=sample),
index=False,
header=True)
self.plot_insert_size(histogram, sample)
df = pd.DataFrame.from_dict(insert, orient='index')
df = df.reset_index()
df = df.rename(columns={'index': 'Samples'})
return df
def raw_reads_content(self):
before_filtering_dataframe = nesteddict()
after_filtering_dataframe = nesteddict()
filtering_result_dataframe = nesteddict()
sequencing_qc = self.read_json()
for sample in sequencing_qc:
try:
before_filtering = sequencing_qc[sample]['summary'][
'before_filtering']
before_filtering = {
'Raw_' + key: before_filtering[key]
for key in before_filtering
}
after_filtering = sequencing_qc[sample]['summary'][
'after_filtering']
after_filtering = {
'Clean_' + key: after_filtering[key]
for key in after_filtering
}
filtering_result = sequencing_qc[sample]['filtering_result']
before_filtering_dataframe[sample] = before_filtering
after_filtering_dataframe[sample] = after_filtering
filtering_result_dataframe[sample] = filtering_result
# quality and content png
read1_after_filtering_quality_curves = sequencing_qc[sample][
'read1_after_filtering']['quality_curves']
read1_after_filtering_content_curves = sequencing_qc[sample][
'read1_after_filtering']['content_curves']
read2_after_filtering_quality_curves = sequencing_qc[sample][
'read2_after_filtering']['quality_curves']
read2_after_filtering_content_curves = sequencing_qc[sample][
'read2_after_filtering']['content_curves']
self.quality_png(read1_after_filtering_quality_curves, 'read1',
sample)
self.content_png(read1_after_filtering_content_curves, 'read1',
sample)
self.quality_png(read2_after_filtering_quality_curves, 'read2',
sample)
self.content_png(read2_after_filtering_content_curves, 'read2',
sample)
except KeyError:
continue
return before_filtering_dataframe, after_filtering_dataframe, filtering_result_dataframe
def read_his(self, file, index):
result = nesteddict()
with open(file, 'rt') as f:
for key, value in enumerate(f):
if key < index:
continue
else:
try:
cov, count = value.strip().split()
result['count'][cov] = count
except ValueError:
break
return result
def dedup_log(self):
dedup = nesteddict()
for sample in self.collect_file:
log = self.collect_file[sample]
# print(log)
line = self.get_line(log, 'LIBRARY') + 1
rec = getline(log, line).strip().split()
UNPAIRED_READS_EXAMINED, READ_PAIRS_EXAMINED, UNMAPPED_READS, PERCENT_DUPLICATION, ESTIMATED_LIBRARY_SIZE = rec[
1], rec[2], rec[4], rec[-2], rec[-1]
#sample[sample]['Unmerged_reads'] = UNPAIRED_READS_EXAMINED
#sample[sample]['Merged_reads'] = READ_PAIRS_EXAMINED
dedup[sample]['Mapped_rate'] = round(
(int(ESTIMATED_LIBRARY_SIZE) - int(UNMAPPED_READS)) /
int(ESTIMATED_LIBRARY_SIZE), 4)
dedup[sample]['PCR_duplication'] = PERCENT_DUPLICATION
#sample[sample]['Library_size'] = ESTIMATED_LIBRARY_SIZE
df = pd.DataFrame.from_dict(dedup, orient='index')
df = df.reset_index()
df = df.rename(columns={'index': 'Samples'})
return df
def read_json(self):
sequencing_qc = nesteddict()
for sample in self.collect_file:
qc = self.json_in(self.collect_file[sample])
sequencing_qc[sample] = qc
return sequencing_qc