def isFiltered(row, clr):
cmd = ' '.join([
'cat',
os.path.join(sf_calls_dir, clr, clr + '-SNP-class_MIS_SNP__.csv'),
os.path.join(sf_calls_dir, clr, clr + '-SNP-class_SYN_SNP__.csv'),
os.path.join(sf_calls_dir, clr, clr + '-SNP-class_LOF_SNP__.csv'), '|',
'grep', row['ind_id'], '|', 'grep', row['CHROM'], '|', 'grep',
str(row['POS'])
])
res = func.runInShell(cmd)
return not bool(res)
def isPossibleDeNovo(row, clr):
if len(row['REF']) == len(row['ALT']):
v_t = 'snp'
else:
v_t = 'indels'
cmd = ' '.join([
'cat',
os.path.join(sf_calls_dir, v_t, clr, row['ind_id']), '|', 'grep',
row['CHROM'], '|', 'grep',
str(row['POS'])
])
res = func.runInShell(cmd)
return not bool(res)
def isDeNovo(row, clr):
if len(row['REF']) == len(row['ALT']):
v_t = 'snp'
v_tt = 'SNP'
else:
v_t = 'indels'
v_tt = 'INDEL'
cmd = ' '.join([
'cat',
os.path.join(sf_calls_dir, v_t, clr, row['batch'],
row['ind_id'] + '-' + v_tt + '-class.csv'), '|', 'grep',
row['ind_id'], '|', 'grep', row['CHROM'], '|', 'grep',
str(row['POS'])
])
res = func.runInShell(cmd)
return not bool(res)
def whyNotDeNovo(row, clr):
cmd = ' '.join([
'cat',
os.path.join(sf_calls_dir, clr, clr + '-SNP-class_ALL_SNP__.csv'), '|',
'grep', row['ind_id'], '|', 'grep', row['CHROM'], '|', 'grep',
str(row['POS'])
])
res = func.runInShell(cmd, True)
# print res
if row['is_dn'] and not row['is_filt']:
res = res.split('\n')[0]
outp = res.split(',')[-len(sf_dn_c_columns):]
else:
outp = [None] * len(sf_dn_c_columns)
ser_out = pandas.Series(outp, sf_dn_c_columns, dtype=str)
# print ser_out
# ser_out = ser_out[sf_dn_c_columns]
return ser_out
dnvo.reset_index(inplace=True)
if dnvo.empty:
sys.exit('No mutations at score %s' % prob_cutoff)
tmp_dir = tempfile.mkdtemp()
print(tmp_dir)
input_file_bn = os.path.splitext(os.path.basename(input_file))[0]
outp_tsv = os.path.join(tmp_dir, input_file_bn + '.tsv')
print(outp_tsv)
func.writePredAsVcf(dnvo, outp_tsv, min_DP=min_DP)
# script_name = os.path.basename(os.path.realpath(sys.argv[0]))
script_name = os.path.abspath(pkg_resources.resource_filename('variants',
'vcf2table.sh'))
cmd = ' '.join([script_name,
outp_tsv,
os.path.dirname(script_name),
input_file_bn,
targ_bed])
print(cmd)
func.runInShell(cmd)
vn = summarizeVariants.summarizeMutations(
os.path.join(tmp_dir,
input_file_bn +
'-ann-onePline.tsv'),
input_file_bn,
output_dir,
config_file)
if rm_tmp == 'yes':
cmd = 'rm -rf %s' % tmp_dir
func.runInShell(cmd)
genome_build = int(cfg['genome_build'])
vep_refseq = cfg['vep_refseq']
if genome_build == 19 or genome_build == 37:
incl_make = '/mnt/xfs1/home/asalomatov/projects/pipeline/ppln/include.mk'
elif int(genome_build) == 38:
incl_make = '/mnt/xfs1/home/asalomatov/projects/pipeline/ppln/include_hg38.mk'
else:
sys.exit('Only builds 19, 37, 38 are supported')
tmp_dir = tempfile.mkdtemp()
print(tmp_dir)
input_file_bn = os.path.splitext(os.path.basename(input_file))[0]
input_lile_dir = os.path.dirname(input_file)
script_name = os.path.abspath(
pkg_resources.resource_filename('variants', 'vcf2tablee.sh'))
script_dir = os.path.dirname(script_name)
cmd = """
vcfintersect -b %(targ_bed)s %(input_file)s > %(tmp_dir)s/%(input_file_bn)s.vcf
echo 'running VEP'
make -f %(script_dir)s/annVEP.mk INCLMK=%(incl_make)s VEPREFSEQ=%(vep_refseq)s PREFIX=%(input_file_bn)s SUFFIX=.vcf INDIR=%(tmp_dir)s OUTDIR=%(tmp_dir)s
"""
print(cmd % locals())
func.runInShell(cmd % locals())
if rm_tmp == 'yes':
cmd = 'rm -rf %s' % tmp_dir
func.runInShell(cmd)
print(tmp_dir)
input_file_bn = os.path.splitext(os.path.basename(input_file))[0]
outp_tsv = os.path.join(tmp_dir, input_file_bn + '.tsv')
print(outp_tsv)
func.writeTableAsVcf(dnvo, outp_tsv)
# script_name = os.path.basename(os.path.realpath(sys.argv[0]))
script_name = os.path.abspath(
pkg_resources.resource_filename('variants', 'vcf2table.sh'))
# 'vcf2table_notarg.sh'))
cmd = ' '.join([
script_name, outp_tsv,
os.path.dirname(script_name), input_file_bn, targ_bed, incl_make,
vep_refseq
])
print(cmd)
func.runInShell(cmd)
vn = summarizeOtherVariants.summarizeMutations(
os.path.join(tmp_dir, input_file_bn + '-ann.vcf.onePline.tsv'),
os.path.join(tmp_dir, input_file_bn + '-vep.tsv'), input_file_bn,
output_dir, config_file)
if rm_tmp == 'yes':
cmd = 'rm -rf %s' % tmp_dir
func.runInShell(cmd)
sys.exit('stop')
def get_spID(x, lab2sp_dict):
if x[:2] == 'SP':
return x
else:
sys.exit(1)
model = os.path.join(model_dir, model[0])
print(model)
m_pkl = joblib.load(model)
list_of_features = m_pkl['features']
# hardcode lvl, this is intended for external use only
lvl = 0
is_keras = bool(int(m_pkl['is_keras']))
if lvl == 0:
m_pkl['extra_col_names'] = []
m_pkl['y_name'] = []
# create output dirs
func.runInShell('mkdir -p ' + output_dir)
if known_vars:
output_dir_known = os.path.join(output_dir, 'known')
func.runInShell('mkdir -p ' + output_dir_known)
# populate ped DF
myped = ped.Ped(ped_file_extended, ['bam', 'vcf'])
f = features.Features(myped, known_vars)
# trio has to be complete with no files missing
if not f.initTrioFor(child_id):
sys.stderr.write('\nfailed to initialize trio for ' + child_id)
sys.exit(1)
else:
sys.stdout.write('\ninitialized trio for ' + child_id)
sys.stdout.write('\n')
