def testSplitStringinTwo(self):
"""
Tests spliting a string into two on | or ,
Return -1, -1 if input is None
Used to split columns of data in summary
"""
result01=split_string_in_two("1|99")
result02=split_string_in_two(None)
self.assertEquals(result01[0], '1')
self.assertEquals(result01[1], '99')
self.assertEquals(result02[0], '-1')
self.assertEquals(result02[1], '-1')
def testSplitStringinTwo(self):
"""
Tests spliting a string into two on | or ,
Return -1, -1 if input is None
Used to split columns of data in summary
"""
result01 = ann.split_string_in_two("1|99")
result02 = ann.split_string_in_two(None)
self.assertEquals(result01[0], '1')
self.assertEquals(result01[1], '99')
self.assertEquals(result02[0], '-1')
self.assertEquals(result02[1], '-1')
def action(args):
(infiles, ) = args.infiles
RefSeqs = {}
if args.RefSeqs:
refs = csv.DictReader(args.RefSeqs, delimiter='\t')
for row in refs:
if row['RefSeq'] :
for transcript in row['RefSeq'].split('/'):
RefSeqs[transcript.split('.')[0]] = transcript
headers = ['Position'] + variant_headers[3:5] + [
'Clinically_Flagged',
'Variant_Type',
'UW_Freq',
'UW_DEC_p',
'Filter',
'1000g_ALL',
'EVS_esp6500_ALL',
'EXAC',
'Gene',
'p.',
'c.',
'Faves_Y/N',
'Ref_Reads',
'Var_Reads',
'Allele_Frac',
'Variant_Phred',
'Cosmic',
'CADD',
'ClinVar',
'Polyphen',
'Sift',
'Mutation_Taster',
'Gerp',
'1000g_AMR',
'1000g_EUR',
'1000g_SAS',
'1000g_EAS',
'1000g_AFR',
'EVS_esp6500_AA',
'EVS_esp6500_EU',
'Transcripts',
'Zygosity',
'Segdup',
'NCI60',
'dbSNP_ID',
'UW_Count',
'GATK_Score',
'ADA_Alter_Splice',
'RF_Alter_Splice',
]
if args.type == 'PINDEL':
headers = ['Position'] + variant_headers[3:5] + [
'Clinically_Flagged',
'Variant_Type',
'UW_Freq',
'Gene',
'p.',
'c.',
'Faves_Y/N',
'Ref_Reads',
'Var_Reads',
'Allele_Frac',
'Transcripts',
'UW_Count',
]
writer = csv.DictWriter(args.outfile,
fieldnames=headers,
quoting=csv.QUOTE_MINIMAL,
extrasaction='ignore',
delimiter='\t')
writer.writeheader()
# accumulate data from all input files for each variant
output = defaultdict(dict)
for fname in infiles:
if args.type == 'SNP':
file_types = snp_file_types
elif args.type == 'INDEL':
file_types = indel_file_types
elif args.type == 'PINDEL':
file_types = pindel_file_types
try:
_, file_type = path.basename(fname).split('.', 1)
except ValueError:
continue
try:
#if the file type matches one we want,
#header ids are output columns
#var_key_ids are chrm:str:stp:ref:var
header_ids, var_key_ids = file_types[file_type]
except KeyError:
if re.search('dropped', fname):
continue
log.warning('no match: %s' % fname)
if args.strict:
sys.exit(1)
continue
for var_key, data in map_headers(fname, header_ids, var_key_ids):
if var_key in output:
output[var_key]=merge_data(output[var_key],data)
else:
output[var_key].update(data)
if output[var_key].has_key('Reads') and not output[var_key].has_key('Var_Reads'):
output[var_key]['Ref_Reads'], output[var_key]['Var_Reads'], output[var_key]['Variant_Phred'] = get_reads(data.get('Read_Headers'),data.get('Reads'))
sort_key = lambda row: [(row[k]) for k in ['chr', 'start', 'stop', 'Ref_Base', 'Var_Base']]
# # write each row (with all data aggregated), modifying fields as necessary
for data in sorted(output.values(), key=sort_key):
variants=[data.get('var_type_2'),data.get('var_type_1')]
data['Variant_Type'] = ','.join(filter(None, variants))
data['Gene'], data['Transcripts'] = munge_gene_and_Transcripts(data, RefSeqs)
data['c.'], data['p.'] = munge_transcript(data, RefSeqs)
data['Polyphen'], data['Sift'],data['Mutation_Taster'],data['Gerp'] = munge_ljb_scores(data)
data['dbSNP_ID'] = data.get('rsid_1') or data.get('rsid_2')
data['1000g_ALL'] = data.get('1000g_ALL') or -1
data['1000g_AMR'] = data.get('1000g_AMR') or -1
data['1000g_SAS'] = data.get('1000g_SAS') or -1
data['1000g_EAS'] = data.get('1000g_EAS') or -1
data['1000g_AFR'] = data.get('1000g_AFR') or -1
data['1000g_EUR'] = data.get('1000g_EUR') or -1
data['UW_DEC_p'] = data.get('UW_DEC_p') or -1
data['EXAC'] = data.get('EXAC').split(',')[0] if data.get('EXAC') else -1
data['EVS_esp6500_ALL'] = data.get('EVS_esp6500_ALL').split(',')[0] if data.get('EVS_esp6500_ALL') else -1
data['EVS_esp6500_AA'] = data.get('EVS_esp6500_AA').split(',')[0] if data.get('EVS_esp6500_AA') else -1
data['EVS_esp6500_EU'] = data.get('EVS_esp6500_EU').split(',')[0] if data.get('EVS_esp6500_EU') else -1
#CADD is raw score, phred score. We only care about phred
_, data['CADD'] = split_string_in_two(data.get('CADD'))
data['ADA_Alter_Splice'],data['RF_Alter_Splice'] = split_string_in_two(data.get('splicing'))
data['UW_Freq'], data['UW_Count'] = split_string_in_two(data.get('UW_Freq_list'))
data['Allele_Frac'] = get_allele_freq(data)
writer.writerow(data)
def action(args):
(infiles, ) = args.infiles
RefSeqs = {}
if args.RefSeqs:
refs = csv.DictReader(args.RefSeqs, delimiter='\t')
for row in refs:
if row['RefSeq']:
for transcript in row['RefSeq'].split('/'):
RefSeqs[transcript.split('.')[0]] = transcript
headers = ['Position'] + variant_headers[3:5] + [
'Clinically_Flagged',
'Variant_Type',
'UW_Freq',
'UW_DEC_p',
'Filter',
'1000g_ALL',
'EVS_esp6500_ALL',
'EXAC',
'Gene',
'p.',
'c.',
'Faves_Y/N',
'Ref_Reads',
'Var_Reads',
'Allele_Frac',
'Variant_Phred',
'Cosmic',
'CADD',
'ClinVar',
'Polyphen',
'Sift',
'Mutation_Taster',
'Gerp',
'1000g_AMR',
'1000g_EUR',
'1000g_SAS',
'1000g_EAS',
'1000g_AFR',
'EVS_esp6500_AA',
'EVS_esp6500_EU',
'Transcripts',
'Zygosity',
'Segdup',
'NCI60',
'dbSNP_ID',
'UW_Count',
'GATK_Score',
'ADA_Alter_Splice',
'RF_Alter_Splice',
]
if args.type == 'PINDEL':
headers = ['Position'] + variant_headers[3:5] + [
'Clinically_Flagged',
'Variant_Type',
'UW_Freq',
'Gene',
'p.',
'c.',
'Faves_Y/N',
'Ref_Reads',
'Var_Reads',
'Allele_Frac',
'Transcripts',
'UW_Count',
]
writer = csv.DictWriter(args.outfile,
fieldnames=headers,
quoting=csv.QUOTE_MINIMAL,
extrasaction='ignore',
delimiter='\t')
writer.writeheader()
# accumulate data from all input files for each variant
output = defaultdict(dict)
for fname in infiles:
if args.type == 'SNP':
file_types = snp_file_types
elif args.type == 'INDEL':
file_types = indel_file_types
elif args.type == 'PINDEL':
file_types = pindel_file_types
try:
_, file_type = path.basename(fname).split('.', 1)
except ValueError:
continue
try:
#if the file type matches one we want,
#header ids are output columns
#var_key_ids are chrm:str:stp:ref:var
header_ids, var_key_ids = file_types[file_type]
except KeyError:
if re.search('dropped', fname):
continue
log.warning('no match: %s' % fname)
if args.strict:
sys.exit(1)
continue
for var_key, data in map_headers(fname, header_ids, var_key_ids):
if var_key in output:
output[var_key] = merge_data(output[var_key], data)
else:
output[var_key].update(data)
if output[var_key].has_key(
'Reads') and not output[var_key].has_key('Var_Reads'):
output[var_key]['Ref_Reads'], output[var_key][
'Var_Reads'], output[var_key][
'Variant_Phred'] = get_reads(
data.get('Read_Headers'), data.get('Reads'))
sort_key = lambda row: [
(row[k]) for k in ['chr', 'start', 'stop', 'Ref_Base', 'Var_Base']
]
# # write each row (with all data aggregated), modifying fields as necessary
for data in sorted(output.values(), key=sort_key):
variants = [data.get('var_type_2'), data.get('var_type_1')]
data['Variant_Type'] = ','.join(filter(None, variants))
data['Gene'], data['Transcripts'] = munge_gene_and_Transcripts(
data, RefSeqs)
data['c.'], data['p.'] = munge_transcript(data, RefSeqs)
data['Polyphen'], data['Sift'], data['Mutation_Taster'], data[
'Gerp'] = munge_ljb_scores(data)
data['dbSNP_ID'] = data.get('rsid_1') or data.get('rsid_2')
data['1000g_ALL'] = data.get('1000g_ALL') or -1
data['1000g_AMR'] = data.get('1000g_AMR') or -1
data['1000g_SAS'] = data.get('1000g_SAS') or -1
data['1000g_EAS'] = data.get('1000g_EAS') or -1
data['1000g_AFR'] = data.get('1000g_AFR') or -1
data['1000g_EUR'] = data.get('1000g_EUR') or -1
data['UW_DEC_p'] = data.get('UW_DEC_p') or -1
data['EXAC'] = data.get('EXAC').split(',')[0] if data.get(
'EXAC') else -1
data['EVS_esp6500_ALL'] = data.get('EVS_esp6500_ALL').split(
',')[0] if data.get('EVS_esp6500_ALL') else -1
data['EVS_esp6500_AA'] = data.get('EVS_esp6500_AA').split(
',')[0] if data.get('EVS_esp6500_AA') else -1
data['EVS_esp6500_EU'] = data.get('EVS_esp6500_EU').split(
',')[0] if data.get('EVS_esp6500_EU') else -1
#CADD is raw score, phred score. We only care about phred
_, data['CADD'] = split_string_in_two(data.get('CADD'))
data['ADA_Alter_Splice'], data[
'RF_Alter_Splice'] = split_string_in_two(data.get('splicing'))
data['UW_Freq'], data['UW_Count'] = split_string_in_two(
data.get('UW_Freq_list'))
data['Allele_Frac'] = get_allele_freq(data)
writer.writerow(data)
def action(args):
(infiles, ) = args.infiles
RefSeqs = {}
if args.RefSeqs:
refs = csv.DictReader(args.RefSeqs, delimiter='\t')
for row in refs:
if row['RefSeq'] :
for transcript in row['RefSeq'].split('/'):
RefSeqs[transcript.split('.')[0]] = transcript
headers = ['Position'] + variant_headers[3:5] + [
'Clinically_Flagged',
'Variant_Type',
'HiSeq_Freq',
'1000g_ALL',
'EVS_esp6500_ALL',
'Gene',
'p.',
'c.',
'Faves_Y/N',
'Ref_Reads',
'Var_Reads',
'Allele_Freq',
'Variant_Phred',
'Cosmic',
'CADD',
'ClinVar',
'Polyphen',
'Sift',
'Mutation_Taster',
'Gerp',
'1000g_AMR',
'1000g_EUR',
'1000g_ASN',
'1000g_AFR',
'EVS_esp6500_AA',
'EVS_esp6500_EU',
'Transcripts',
'Zygosity',
'Segdup',
'NCI60',
'dbSNP_ID',
'HiSeq_Count',
'MiSeq_Freq',
'MiSeq_Count',
'GATK_Score'
]
# accumulate data from all input files for each variant
output = defaultdict(dict)
for fname in infiles:
try:
_, file_type = path.basename(fname).split('.', 1)
except ValueError:
continue
try:
#if the file type matches one we want,
#header ids are output columns
#var_key_ids are chrm:str:stp:ref:var
header_ids, var_key_ids = file_types[file_type]
except KeyError:
if re.search('dropped', fname):
log.warning('no match: %s' % fname)
if args.strict:
sys.exit(1)
continue
for var_key, data in map_headers(fname, header_ids, var_key_ids):
output[var_key].update(data)
writer = csv.DictWriter(args.outfile,
fieldnames=headers,
quoting=csv.QUOTE_MINIMAL,
extrasaction='ignore',
delimiter='\t')
writer.writeheader()
sort_key = lambda row: [(row[k]) for k in ['chr', 'start', 'stop']]
# write each row (with all data aggregated), modifying fields as necessary
for data in sorted(output.values(), key=sort_key):
# # modify any specific fields here
data['Variant_Type'] = data.get('var_type_2') if data.get('var_type_2', '').strip() else data.get('var_type_1')
data['Gene'], data['Transcripts'] = munge_gene_and_Transcripts(data, RefSeqs)
data['c.'], data['p.'] = munge_transcript(data, RefSeqs)
data['dbSNP_ID'] = data.get('rsid_1') or data.get('rsid_2')
data['1000g_ALL'] = data.get('1000g_ALL') or -1
data['1000g_AMR'] = data.get('1000g_AMR') or -1
data['1000g_ASN'] = data.get('1000g_ASN') or -1
data['1000g_AFR'] = data.get('1000g_AFR') or -1
data['1000g_EUR'] = data.get('1000g_EUR') or -1
data['EVS_esp6500_ALL'] = data.get('EVS_esp6500_ALL') or -1
data['EVS_esp6500_AA'] = data.get('EVS_esp6500_AA') or -1
data['EVS_esp6500_EU'] = data.get('EVS_esp6500_EU') or -1
#CADD is raw score, phred score. We only care about phred
_, data['CADD'] = split_string_in_two(data.get('CADD'))
data['Ref_Reads'], data['Var_Reads'], data['Variant_Phred'] = get_reads(data.get('Reads'))
data['MiSeq_Freq'], data['MiSeq_Count'] = split_string_in_two(data.get('Mi_Freq_list'))
data['HiSeq_Freq'], data['HiSeq_Count'] = split_string_in_two(data.get('Hi_Freq_list'))
data['Allele_Freq'] = get_allele_freq(data)
writer.writerow(data)