def flatten_vcf_data(fields, vcfdata, flatdata=defaultdict(dict)):
infofields = fields[:9]
samples = fields[9:]
for d in vcfdata:
if d['CHROM']=='chr7':
(d['CHROM'], pos) = cftr.hg19_to_CFTR(d['CHROM'], d['POS'])
d['POS'] = str(pos)
poskey = create_pos_key(d['CHROM'], d['POS'])
ids = d['ID'].split(';')
dbsnp = [ rs for rs in ids if rs.startswith('rs') ]
d['hg19_coordinates'] = "{}:{}".format(*cftr.CFTR_to_hg19(
d['CHROM'], d['POS']))
d['ID'] = ';'.join(dbsnp) if dbsnp else '-'
d['hg19_ID'] = '-'
for sample in samples:
sfields = d['FORMAT'].split(':')
sdata = dict(zip(sfields, d[sample].split(':')))
idata = dict([ v.split('=') for v in d['INFO'].split(';') \
if '=' in v ])
if sdata['GT']=='0/0' or sdata['GT']=='./.':
continue
sample_row = d.copy()
sample_row['Sample ID'] = sample
gtvals = [int(i) for i in sdata['GT'].split('/')]
altchoices = [d['REF'],] + d['ALT'].split(',')
altvals = set([ altchoices[i] for i in gtvals if i>0 ])
sample_row['ALT'] = ','.join(altvals)
sample_row['hom/het'] = 'hom' if gtvals[0]==gtvals[1] else 'het'
(sample_row['DP'], sample_row['AF']) = DPcalculate_AF(sdata,
d['INFO'])
sample_row['QD'] = idata['QD'] if 'QD' in idata else ''
sample_row['FS'] = idata['FS'] if 'FS' in idata else ''
newgtvals = []
if gtvals[0]==0:
sdata['GT'] = '0/1'
elif gtvals[0]==gtvals[1]:
sdata['GT'] = '1/1'
else:
sdata['GT'] = '1/2'
sample_row['format_values'] = ':'.join([ sdata[f] for f in sfields ])
flatdata[sample][poskey] = sample_row
newfields = ['Sample ID',]+infofields+['format_values', 'hom/het',
'ROI (Y/N)', 'c./p.(AnnoVar)', 'DP', 'AF', 'QD', 'FS',
'hg19_coordinates', ]
return (flatdata, newfields)
def get_annovar_data(annovar_files):
if not annovar_files: return {}
sys.stderr.write("\nParsing annovar data {}\n".format(annovar_files))
afiles = []
for ext in ('.variant_function', '.exonic_variant_function'):
files = [ f for f in annovar_files if f.endswith(ext) ]
if files:
afiles.append(files[0])
if not os.path.isfile(files[0]):
sys.stderr.write(" Not a valid file: {}\n".format(files[0]))
sys.exit(1)
else:
sys.stderr.write(" Could not find annovar {} file\n".format(ext))
sys.exit(1)
sys.stderr.write(" Reading {}\n".format(afiles[0]))
with open(afiles[0], 'r') as fh:
data_rows = [ [l.rstrip().split("\t"),] for l in fh.readlines() \
if len(l)>1 ]
sys.stderr.write(" Read {} lines\n".format(len(data_rows)))
sys.stderr.write(" Reading {}\n".format(afiles[1]))
with open(afiles[1], 'r') as fh:
numdat = 0
for l in fh.readlines():
d = l.rstrip().split("\t")
if not d: continue
linenum = int(d.pop(0).lstrip('line'))
data_rows[linenum-1].append(d)
numdat += 1
sys.stderr.write(" Read {} lines\n".format(numdat))
annovar_data = defaultdict(dict)
for row in data_rows:
if len(row[0])>10:
(chrom_hg19, pos_hg19) = row[0][10:12]
else:
(chrom_hg19, pos_hg19) = row[0][2:4]
(chrom, pos) = cftr.hg19_to_CFTR(chrom_hg19, pos_hg19)
poskey = create_pos_key(chrom, pos)
if len(row)>1: #has exonic_variant_function data
annovar_data[poskey][row[1][1].rstrip(',')] = row[1][0]
elif ':' in row[0][1]:
annovar_data[poskey][row[0][1]] = row[0][0]
return annovar_data
def flatten_vcf_data(fields, vcfdata, flatdata=defaultdict(dict)):
infofields = fields[:9]
samples = fields[9:]
ifields = [] #'DP', 'SRF', 'SRR', 'SAF', 'SAR','ABP', 'QA', 'AO']
for d in vcfdata:
if d['CHROM']=='chr7':
(d['CHROM'], pos) = cftr.hg19_to_CFTR(d['CHROM'], d['POS'])
d['POS'] = str(pos)
poskey = create_pos_key(d['CHROM'], d['POS'])
ids = d['ID'].split(';')
dbsnp = [ rs for rs in ids if rs.startswith('rs') ]
d['hg19_coordinates'] = "{}:{}".format(*cftr.CFTR_to_hg19(
d['CHROM'], d['POS']))
d['ID'] = ';'.join(dbsnp) if dbsnp else '-'
d['hg19_ID'] = '-'
for sample in samples:
sfields = d['FORMAT'].split(':')
sdata = dict(zip(sfields, d[sample].split(':')))
idata = dict([ v.split('=') for v in d['INFO'].split(';') \
if '=' in v ])
if sdata['GT']=='0/0' or sdata['GT']=='./.':
continue
sample_row = d.copy()
sample_row['Sample ID'] = sample
gtvals = [int(i) for i in sdata['GT'].split('/')]
sample_row['hom/het'] = 'hom' if gtvals[0]==gtvals[1] else 'het'
(sample_row['alt_AF'], sample_row['DP']) = calculate_AF(idata)
# for f in ifields: sample_row[f] = idata[f]
add_avg_base_quality(sample_row, idata)
sample_row['format_values'] = ':'.join([ sdata[f] for f in sfields ])
flatdata[sample][poskey] = sample_row
ifields.extend(['QA/AO',])
newfields = ['Sample ID',]+infofields+['format_values', 'hom/het',
'ROI (Y/N)', 'c./p.(AnnoVar)', 'DP', 'alt_AF',]+\
ifields + ['hg19_coordinates', ]
return (flatdata, newfields)