def run():
import argparse
parser = argparse.ArgumentParser(usage=__usage__)
# Necessary arguments
parser.add_argument("-v", "--vcf", required=True)
parser.add_argument("-f", "--fam", required=True)
# Optional arguments
parser.add_argument("-g",
"--gen",
required=False,
default="hg38",
choices=["hg19", "hg38"])
parser.add_argument("-i", "--info", required=False)
args = parser.parse_args()
vcf_filepath = args.vcf
fam_filepath = args.fam
gen = args.gen
info_keys = []
if args.info:
f = open(args.info, "r")
for line in f:
info_keys.append(line.rstrip())
else:
info_keys = [
"VQSLOD", "ClippingRankSum", "BaseQRankSum", "FS", "SOR", "MQ",
"MQRankSum", "QD", "ReadPosRankSum"
]
from vcf import parse
parse(vcf_filepath, fam_filepath, info_keys=info_keys)
def vcf_file(filepath=None, input=None, delim=",", quote='"'):
if filepath is not None:
input = open(filepath, 'rb')
csv_input = csv.reader(input, delimiter=delim, quotechar=quote)
# for each vc_group
for row in csv_input:
row = [None if f == '' else f for f in row]
info = vcf.parse('info', row[35 - 1])
vc_group_columns = {
'chromosome' : row[22 - 1],
'start_posn' : row[23 - 1],
'end_posn' : row[24 - 1],
'ref' : vcf.parse('ref', row[25 - 1]),
'dbsnp_id' : vcf.parse('dbsnp_id', row[30 - 1]),
# 'genotype_format' : row[36 - 1],
'quality' : row[33 - 1],
'filter' : row[34 - 1],
# annovar columns
'otherinfo' : row[27 - 1],
'func' : row[1 - 1],
'gene' : row[2 - 1],
'exonicfunc' : row[3 - 1],
'aachange' : row[4 - 1],
'conserved' : row[5 - 1],
'1000g2011may_all' : row[8 - 1],
'dbsnp135' : row[9 - 1],
'ljb_phylop_pred' : row[12 - 1],
'ljb_sift_pred' : row[14 - 1],
'ljb_polyphen2_pred' : row[16 - 1],
'ljb_lrt_pred' : row[18 - 1],
'ljb_mutationtaster_pred' : row[20 - 1],
'ljb_gerppp' : row[21 - 1],
'segdup' : row[6 - 1],
'esp5400_all' : row[7 - 1],
'avsift' : row[10 - 1],
'ljb_phylop' : row[11 - 1],
'ljb_sift' : row[13 - 1],
'ljb_polyphen2' : row[15 - 1],
'ljb_lrt' : row[17 - 1],
'ljb_mutationtaster' : row[19 - 1],
# vc_group_info columns
# 'info_source' : row[36 - 1],
'ds' : info.get('DS', False),
'inbreeding_coeff' : info.get('InbreedingCoeff'),
'base_q_rank_sum' : info.get('BaseQRankSum'),
'mq_rank_sum' : info.get('MQRankSum'),
'read_pos_rank_sum' : info.get('ReadPosRankSum'),
'dels' : info.get('Dels'),
'fs' : info.get('FS'),
'haplotype_score' : info.get('HaplotypeScore'),
'mq' : info.get('MQ'),
'qd' : info.get('QD'),
'sb' : info.get('SB'),
'vqslod' : info.get('VQSLOD'),
'an' : info.get('AN'),
'dp' : info.get('DP'),
'mq0' : info.get('MQ0'),
'culprit' : info.get('culprit'),
}
vc_group_table = vc_group(vc_group_columns)
alts = vcf.parse('alts', row[32 - 1])
# for each vc_group_allele in (vc x alt alleles in vc_group)
vc_group_allele_fields = [
alts,
# vc_group_allele_info
get_list(info, 'AF'),
get_list(info, 'MLEAF'),
get_list(info, 'AC'),
get_list(info, 'MLEAC'),
]
vc_group_allele_columns = [{
# 'vc_group_id' : vc_group_id,
'allele' : allele,
'af' : af,
'mle_af' : mle_af,
'ac' : ac,
'mle_ac' : mle_ac,
} for allele, af, mle_af, ac, mle_ac in arity_zip(vc_group_allele_fields, table='vc_group', key="alt alleles in vc_group")]
add_columns(vc_group_allele_columns, vc_group_table.vc_group_allele)
ref_and_alts = as_list(vc_group_columns['ref']) + alts
# for each vc in vc_group
for genotype in [vcf.parse('genotype', row[gf]) for gf in xrange(37 - 1, 48)]:
# vc_columns['genotype_source'] = row[gf]
vc_columns = {
# 'vc_group_id' : vc_group_table.lastrowid,
'zygosity' : row[27 - 1],
}
patient_columns = {
}
# vc_columns['patient_id'] = patient_table.lastrowid
if not (type(genotype) == tuple and genotype[0] == (None, None)):
((allele1_idx, allele2_idx), vc_columns['phased']) = genotype['GT']
vc_columns['allele1'] = ref_and_alts[allele1_idx]
vc_columns['allele2'] = ref_and_alts[allele2_idx]
vc_columns['read_depth'] = genotype.get('DP')
vc_columns['genotype_quality'] = genotype.get('GQ')
vc_table = vc(vc_columns)
vc_group_table.vc.append(vc_table)
# for each vc_genotype in (alleles in vc_group x alleles in vc_group x vc)
vc_genotype_fields = [
vcf.ordered_alleles(vc_group_columns['ref'], alts),
as_list(genotype.get('PL')),
]
vc_genotype_columns = [{
# 'vc_id': vc_id,
'allele1': vc_genotype_allele1,
'allele2': vc_genotype_allele2,
'phred_likelihood': phred_likelihood,
} for (vc_genotype_allele1, vc_genotype_allele2), phred_likelihood in arity_zip(vc_genotype_fields, table='vc_genotype', key="biallelic genotypes in vc_group")]
add_columns(vc_genotype_columns, vc_table.vc_genotype)
# for each vc_allele in (vc x alleles in ref, alts)
vc_allele_fields = [
ref_and_alts,
get_list(genotype, 'AD'),
]
vc_allele_columns = [{
# 'vc_id': vc_id,
'allele': allele,
'allelic_depth': allelic_depth,
} for allele, allelic_depth in arity_zip(vc_allele_fields, table='vc_allele', key="ref and alt alleles in vc_group")]
add_columns(vc_allele_columns, vc_table.vc_allele)
yield vc_group_table
input.close()
def load_genome_summary(db, input, delim=",", quote='"', dry_run=False, records=None, quiet=False, autocommit=False):
# this script will run properly on InnoDB engine without autocommit; sadly, such is not the case for NDB, where we get
# the error:
# Got temporary error 233 'Out of operation records in transaction coordinator (increase MaxNoOfConcurrentOperations)' from NDBCLUSTER
# db.autocommit(True)
def insert_wrapper(table, dic_or_dics, do_insert):
if not dry_run:
# try:
return do_insert(dic_or_dics)
# except Exception as e:
# msg = e.message if e.message else e.__str__()
# raise type(e)(msg + " at line {lineno}".format(lineno=1))
if not quiet:
print "insert into {table} {dic_or_dics}".format(table=table.name, dic_or_dics=dic_or_dics)
def insert_many(table, values):
insert_wrapper(table, values, lambda vs: table.insert_many(values=vs))
def insert(table, dic):
insert_wrapper(table, dic, lambda d: table.insert(d))
def arity_zip(args, error=None, table=None, key=None):
if error is None:
error = "Number of {table} columns don't all match the number of {key}; " + \
"skipping insertion into {table} at line {lineno}"
return check_arity_zip(args, error.format(lineno=1, table=table.name, key=key))
c = db.cursor()
if autocommit:
c.execute("""SET autocommit = 1;""")
else:
c.execute("""SET autocommit = 0;""")
csv_input = csv.reader(input, delimiter=delim, quotechar=quote)
vc_group_table = sql.table.oursql('vc_group', cursor=c)
vc_table = sql.table.oursql('vc', cursor=c)
patient_table = sql.table.oursql('patient', cursor=c)
# table we can batch insert (i.e. tables whose lastrowid's we do not need)
vc_group_allele_table = sql.table.oursql('vc_group_allele', cursor=c, fields=['vc_group_id', 'allele', 'af', 'mle_af', 'ac', 'mle_ac'])
vc_genotype_table = sql.table.oursql('vc_genotype', cursor=c, fields=['vc_id', 'allele1', 'allele2', 'phred_likelihood'])
vc_allele_table = sql.table.oursql('vc_allele', cursor=c, fields=['vc_id', 'allele', 'allelic_depth'])
# for each vc_group
for row in csv_input:
row = [None if f == '' else f for f in row]
info = vcf.parse('info', row[35 - 1])
vc_group_columns = {
'chromosome' : row[22 - 1],
'start_posn' : row[23 - 1],
'end_posn' : row[24 - 1],
'ref' : vcf.parse('ref', row[25 - 1]),
'dbsnp_id' : vcf.parse('dbsnp_id', row[30 - 1]),
# 'genotype_format' : row[36 - 1],
'quality' : row[33 - 1],
'filter' : row[34 - 1],
# annovar columns
'otherinfo' : row[27 - 1],
'func' : row[1 - 1],
'gene' : row[2 - 1],
'exonicfunc' : row[3 - 1],
'aachange' : row[4 - 1],
'conserved' : row[5 - 1],
'1000g2011may_all' : row[8 - 1],
'dbsnp135' : row[9 - 1],
'ljb_phylop_pred' : row[12 - 1],
'ljb_sift_pred' : row[14 - 1],
'ljb_polyphen2_pred' : row[16 - 1],
'ljb_lrt_pred' : row[18 - 1],
'ljb_mutationtaster_pred' : row[20 - 1],
'ljb_gerppp' : row[21 - 1],
'segdup' : row[6 - 1],
'esp5400_all' : row[7 - 1],
'avsift' : row[10 - 1],
'ljb_phylop' : row[11 - 1],
'ljb_sift' : row[13 - 1],
'ljb_polyphen2' : row[15 - 1],
'ljb_lrt' : row[17 - 1],
'ljb_mutationtaster' : row[19 - 1],
# vc_group_info columns
# 'info_source' : row[36 - 1],
'ds' : info.get('DS', False),
'inbreeding_coeff' : info.get('InbreedingCoeff'),
'base_q_rank_sum' : info.get('BaseQRankSum'),
'mq_rank_sum' : info.get('MQRankSum'),
'read_pos_rank_sum' : info.get('ReadPosRankSum'),
'dels' : info.get('Dels'),
'fs' : info.get('FS'),
'haplotype_score' : info.get('HaplotypeScore'),
'mq' : info.get('MQ'),
'qd' : info.get('QD'),
'sb' : info.get('SB'),
'vqslod' : info.get('VQSLOD'),
'an' : info.get('AN'),
'dp' : info.get('DP'),
'mq0' : info.get('MQ0'),
'culprit' : info.get('culprit'),
}
insert(vc_group_table, vc_group_columns)
alts = vcf.parse('alts', row[32 - 1])
# for each vc_group_allele in (vc x alt alleles in vc_group)
vc_group_allele_fields = [
alts,
# vc_group_allele_info
get_list(info, 'AF'),
get_list(info, 'MLEAF'),
get_list(info, 'AC'),
get_list(info, 'MLEAC'),
]
insert_many(vc_group_allele_table, values=[[ vc_group_table.lastrowid, allele, af, mle_af, ac, mle_ac ]
for allele, af, mle_af, ac, mle_ac in arity_zip(vc_group_allele_fields, table=vc_group_table, key="alt alleles in vc_group")])
vc_columns = {
'vc_group_id' : vc_group_table.lastrowid,
'zygosity' : row[27 - 1],
}
ref_and_alts = as_list(vc_group_columns['ref']) + alts
# for each vc in vc_group
for genotype in [vcf.parse('genotype', row[gf]) for gf in xrange(37 - 1, 48)]:
# vc_columns['genotype_source'] = row[gf]
patient_columns = {
}
insert(patient_table, patient_columns)
vc_columns['patient_id'] = patient_table.lastrowid
if not (type(genotype) == tuple and genotype[0] == (None, None)):
((allele1_idx, allele2_idx), vc_columns['phased']) = genotype['GT']
vc_columns['allele1'] = ref_and_alts[allele1_idx]
vc_columns['allele2'] = ref_and_alts[allele2_idx]
vc_columns['read_depth'] = genotype.get('DP')
vc_columns['genotype_quality'] = genotype.get('GQ')
insert(vc_table, vc_columns)
# for each vc_genotype in (alleles in vc_group x alleles in vc_group x vc)
vc_genotype_fields = [
vcf.ordered_alleles(vc_group_columns['ref'], alts),
as_list(genotype.get('PL')),
]
insert_many(vc_genotype_table, values=[[vc_table.lastrowid, vc_genotype_allele1, vc_genotype_allele2, phred_likelihood]
for (vc_genotype_allele1, vc_genotype_allele2), phred_likelihood in arity_zip(vc_genotype_fields, table=vc_genotype_table, key="biallelic genotypes in vc_group")])
# for each vc_allele in (vc x alleles in ref, alts)
vc_allele_fields = [
ref_and_alts,
get_list(genotype, 'AD'),
]
insert_many(vc_allele_table, values=[[vc_table.lastrowid, allele, allelic_depth]
for allele, allelic_depth in arity_zip(vc_allele_fields, table=vc_allele_table, key="ref and alt alleles in vc_group")])
vc_group_allele_table.flush_buffer()
vc_genotype_table.flush_buffer()
vc_allele_table.flush_buffer()
db.commit()
c.close()
def main():
breakends = vcf.parse('../data/test.vcf', '../data/toto.lengths')
print breakends
print breakends.avg()
breakends.validate()
def main():
breakends = vcf.parse('../data/test.vcf', '../data/toto.lengths')
print breakends
print breakends.avg()
breakends.validate()
