def compare_db_to_file(file):
stat = Stat()
gnomAD = GnomAD()
with open(file) as input:
while (True):
line = input.readline()
if (not line):
break
v = Variant(line, gnomAD_connection=gnomAD)
alt_list = v.alt_list()
x1 = 0
af2 = v.get_gnomad_af()
for alt in alt_list:
af1 = gnomAD.get_af(v.chr_num(), v.lowest_coord(), v.ref(),
alt)
if (af1):
x1 += 1
if (af1 == None and af2 == None):
stat.f()
elif (af1 == af2):
stat.t()
elif (af1):
if (af2 == None):
stat.one()
else:
stat.diff()
else:
stat.two()
if (af2 and (x1 == 0)):
stat.two0()
return stat
def output_raw_calls(infile, outfile, limit=None, start=1):
n_out = limit / 20 if limit > 0 else 100
n = 0
l = 0
result = list()
with open(infile) as source, open(outfile, "w") as destination:
while (True):
line = source.readline()
if (not line):
break
l += 1
if (l < start):
continue
v = Variant(line)
n += 1
if (n % n_out == 0):
print n
alleles = v.alt_list()
for allele in alleles:
variant = dict()
variant["chromosome"] = v.chr_num()
variant["position"] = v.start()
variant["reference"] = v.ref()
variant["alternative"] = allele
result.append(variant)
json.dump(result, destination)
print "Variants processed: {}".format(n)
def get_anfisa_json(self, chromosome, start, end, allele):
region = "{}:{}:{}".format(chromosome, start, end)
client = EnsemblRestClient()
variants = client.get_consequences(region, allele)
if (not variants or len(variants) == 0):
return None
records = [
Variant(v, connectors=self.connectors).get_view_json()
for v in variants
]
return records
def annotate(workspace):
f = workspace["file"]
metadata = get_md(f)
if (metadata):
version = Variant.get_version()
old_version = metadata["versions"]["annotations"]
if (old_version == version):
print "Case is already annotated with the same version"
return
case = workspace["mongo-name"]
fname = os.path.basename(f)
casedir = os.path.dirname(f)
path = casedir.split('/')
case_full_id = None
for p in path:
if (case in p):
case_full_id = p
break
if (not case_full_id):
case_full_id = "{}_wgs".format(case)
if (not case in fname):
print "Skipping non-standard case: {}".format(fname)
return
remote = "{aws_user}@{annotation_server}".format(
annotation_server=annotation_server, aws_user=aws_user)
remote_dir = "/data/bgm/cases/{}".format(case)
cmd = "export PYTHONPATH=/data/bgm/anfisa ; cd {remote_dir} ; python -m annotations.annotator ".format(
remote_dir=remote_dir)
cmd = "{base} -i {case_id}_seq_a_boo_regions.vep.json ".format(
base=cmd, case_id=case_full_id)
print "Running {cmd} on ${annotation_server}".format(
cmd=cmd, annotation_server=annotation_server)
ssh = 'ssh -t {remote} "{cmd}"'.format(cmd=cmd, remote=remote)
print ssh
ret = os.system(ssh)
if (ret):
raise Exception("Returned {}".format(ret))
result = "{}/{}_anfisa.json".format(remote_dir, case)
scp = "scp {remote}:{result} {local}".format(remote=remote,
local=f,
result=result)
print scp
ret = os.system(scp)
if (ret):
raise Exception("Returned {}".format(ret))
with open(f) as input, open(out, "w") as out1, HGMD() as hgmd, \
GnomAD() as gnomAD, \
GTF() as gtf, \
SpliceAI() as spliceAI, \
ClinVar() as clinvar:
cns = {
"hgmd": hgmd,
"gnomAD": gnomAD,
"spliceAI": spliceAI,
"liftover": hg19_to_38_converter,
"clinvar": clinvar,
"gtf": gtf.prepare_lookup(transcript=True),
"beacon": beacon
}
metadata = Variant.get_metadata(vcf_header=vcf_header,
samples=samples,
case=case)
print metadata["versions"]
out1.write(json.dumps(metadata) + '\n')
while (True):
line = input.readline()
if (not line):
break
l += 1
if (l < start):
continue
try:
v = Variant(line,
vcf_header=vcf_header,
samples=samples,
case=case,