def main():
"""
This wrapper script will run the tool mutect within the mutect docker
container for the precision immuno project. The wrapper requires
1. mutect
2. java (For running mutect)
3. twoBitToFa from the kent tools library (For extracting the reference
genome in case indexing is required)
4. lftp for downloading the cosmic vcf
Unless specified, the program will look for default executables on $PATH.
The program DOES NOT look for jar files and they are required to be
passed during execution.
"""
# Parse the arguments using prepare.parse_args()
params = prepare.parse_args(main.__doc__, 'mutect', 'mutect_calls')
# params ERROR handling
if not (params.java_Xmx.endswith('G') or params.java_Xmx.endswith('M')):
raise pi_errors.ParameterError(
dt.now().strftime('%I:%M %p %b %d, %Y') + \
': Please use a suitable value for --Xmx.', params.logfile)
params.java_executable = pi_errors.test_param_value(params.java_executable,
'java',
'--java',
params.logfile)
params.mutect_jar = pi_errors.test_param_value(params.mutect_jar,
'Mutect jar',
'--mutect_jar',
params.logfile)
# If Indexing is required, does twoBitToFa point to a valid file?
if params.index_location is None:
params.tbtf_executable = pi_errors.test_param_value(
params.tbtf_executable, 'twoBitToFa', '--twoBitToFa',
params.logfile)
# Do the dnsnp and cosmic vcfs exist?
if params.dbsnp_file == 'DOWNLOAD' or params.cosmic_file == 'DOWNLOAD':
# First ensure the vcf storage location has been provided
if params.vcf_location is None:
raise pi_errors.ParameterError(
dt.now().strftime('%I:%M %p %b %d, %Y') + \
': --vcf_location cannot be empty if either --cosmic, ' + \
'--dbsnp, or --genome_fasta are empty.', params.logfile)
else:
params.vcf_location = os.path.abspath(params.vcf_location)
# Download dbsnp file if required
if params.dbsnp_file == 'DOWNLOAD':
if os.path.exists('/'.join([params.vcf_location, '00-All.vcf'])):
params.dbsnp_file = '/'.join([params.vcf_location,
'00-All.vcf'])
else:
params.dbsnp_file = prepare.download_vcf('dbsnp', params)
# Download cosmic file if required
if params.cosmic_file == 'DOWNLOAD':
if os.path.exists('/'.join([params.vcf_location,
'Cosmic_sorted.vcf'])):
params.cosmic_file = '/'.join([params.vcf_location,
'Cosmic_sorted.vcf'])
else:
params.cosmic_file = prepare.download_vcf('cosmic', params)
# Download genome fasta if required
if params.genome_fasta == 'DOWNLOAD':
if params.vcf_location is None:
# If params.vcf_location is None, set it to the output directory
params.vcf_location = params.outdir
# Does the fasta exist in the vcf_location directory?
if os.path.exists(''.join([params.vcf_location, '/',
params.genome_version, '.fa'])):
params.genome_fasta = ''.join([params.vcf_location, '/',
params.genome_version, '.fa'])
else:
params.genome_fasta = prepare.get_genome(params.genome_version,
params.vcf_location,
params.tbtf_executable,
params.logfile)
else:
params.genome_fasta = pi_errors.test_param_value(params.genome_fasta,
'Genomic Fasta',
'--genome_fasta',
params.logfile)
# Move to working directory before doing I/O intensive work
os.chdir(params.working_dir)
# Call the program
mutect_call = [params.java_executable, ''.join(['-Xmx', params.java_Xmx]),
'-jar'] # Base java call
mutect_call.append(params.mutect_jar)
mutect_call.extend(['-T', 'MuTect'])
mutect_call.extend(['-R', params.genome_fasta])
mutect_call.extend(['--cosmic', params.cosmic_file])
mutect_call.extend(['--dbsnp', params.dbsnp_file])
mutect_call.extend(['--input_file:normal', params.norm_d_file])
mutect_call.extend(['--input_file:tumor', params.tum_d_file])
mutect_call.extend(['--out', ''.join([params.out_prefix, '.out'])])
return_value = call(mutect_call)
if return_value != 0:
raise pi_errors.MyRuntimeError(
dt.now().strftime('%I:%M %p %b %d, %Y') + \
': MuTect failed.', params.logfile)
with open(''.join([params.out_prefix, '.out']), 'r') as mutect_file, \
open(''.join([params.out_prefix, 'non_rejected.out']), 'w') as \
nr_file:
for line in mutect_file:
line = line.strip()
if line.startswith('#'):
print(line, file=nr_file)
continue
if line.startswith('contig'):
print('#', line, sep='', file=nr_file)
continue
line = line.split('\t')
if line[50] == 'REJECT':
continue
else:
print(line, sep='\t', file=nr_file)
print('PROGRESS ' + dt.now().strftime('%I:%M %p %b %d, %Y') + ': ' +
'Mutect run completed. Finishing up...', file=params.logfile)
# Move files from temp directory to outdir
prepare.move_output(params)
print('RESULT ' + dt.now().strftime('%I:%M %p %b %d, %Y') + ': Process ' +
'completed', file=params.logfile)
params.logfile.close()
def process_parameters(params):
'''
This module conducts the error handling for all parmeters passed to the
program.
'''
# Does the provided radia binary provided exist?
params.radia_executable = pi_errors.test_param_value(
params.radia_executable, 'radia', '--radia', params.logfile)
# Setup filterRadia.py
params.filter_radia_executable = '/'.join([os.path.split(
params.radia_executable)[0], 'filterRadia.py'])
params.filter_radia_executable = pi_errors.test_param_value(
params.filter_radia_executable, 'filterradia', '--radia',
params.logfile)
# Test input files
params.tum_d_file = pi_errors.test_param_value(params.tum_d_file,
'Tumor DNA',
'--tum_dna_file',
params.logfile)
params.norm_d_file = pi_errors.test_param_value(params.norm_d_file,
'Normal DNA',
'--norm_dna_file',
params.logfile)
if params.tum_r_file is not None:
params.tum_r_file = pi_errors.test_param_value(params.tum_r_file,
'Tumor RNA',
'--tum_rna_file',
params.logfile)
# If you don't have a reference, you need twoBitToFasta
if params.index_location is None:
params.tbtf_executable = pi_errors.test_param_value(
params.tbtf_executable, 'twoBitToFa', '--twoBitToFa',
params.logfile)
# Are dnsnp or cosmic vcf required?
if params.dbsnp_file == 'DOWNLOAD' or params.cosmic_file == 'DOWNLOAD' or \
params.genome_fasta == 'DOWNLOAD':
# Ensure the vcf storage location has been provided
if params.vcf_location is None:
raise pi_errors.ParameterError(
dt.now().strftime('%I:%M %p %b %d, %Y') + \
': --vcf_location cannot be empty if either --cosmic, ' + \
'--dbsnp, or --genome_fasta are empty.', params.logfile)
else:
params.vcf_location = os.path.abspath(params.vcf_location)
# Download dbsnp file if required
if params.dbsnp_file == 'DOWNLOAD':
if os.path.exists('/'.join([params.vcf_location, '00-All.vcf'])):
params.dbsnp_file = '/'.join([params.vcf_location,
'00-All.vcf'])
else:
params.dbsnp_file = prepare.download_vcf('dbsnp', params)
# Download cosmic file if required
if params.cosmic_file == 'DOWNLOAD':
if os.path.exists('/'.join([params.vcf_location,
'Cosmic_sorted.vcf'])):
params.cosmic_file = '/'.join([params.vcf_location,
'Cosmic_sorted.vcf'])
else:
params.cosmic_file = prepare.download_vcf('cosmic', params)
# Download genome fasta if required
if params.genome_fasta == 'DOWNLOAD' or not \
os.path.exists(params.genome_fasta):
if os.path.exists(''.join([params.vcf_location, '/',
params.genome_version, '.fa'])):
params.genome_fasta = ''.join([params.vcf_location, '/',
params.genome_version, '.fa'])
else:
params.genome_fasta = prepare.get_genome(
params.genome_version, params.vcf_location,
params.twoBitToFa_executable, sys.stderr)
else:
params.genome_fasta = os.path.abspath(params.genome_fasta)
# Set up the value for rna_fasta
if params.rna_fasta == 'GENOME_FASTA':
params.rna_fasta = params.genome_fasta
else:
params.rna_fasta = pi_errors.test_param_value(params.rna_fasta,
'RNA Fasta',
'--rna_fasta',
params.logfile)
# Ensure the other databases are set up correctly
# The package path is 2 levels above the
radia_pkg_path = os.path.split(os.path.split(params.radia_executable)[0])[0]
database_map = defaultdict()
test_database(params.blacklist, 'blacklist', radia_pkg_path,
params.vcf_location, 'data/hg19/blacklists/1000Genomes/' + \
'phase1/', database_map)
test_database(params.retrogenes, 'retrogenes', radia_pkg_path,
params.vcf_location, 'data/hg19/retroGenes/', database_map)
test_database(params.pseudogenes, 'pseudogenes', radia_pkg_path,
params.vcf_location, 'data/hg19/peudoGenes/', database_map)
test_database(params.broad_targets, 'broad_targets', radia_pkg_path,
params.vcf_location, 'data/hg19/broadTargets/',
database_map)
test_database(params.rna_blacklist, 'rna_blacklist', radia_pkg_path,
params.vcf_location, 'data/rnaGeneBlacklist.tab',
database_map)
test_database(params.rna_family_blacklist, 'rna_family_blacklist',
radia_pkg_path, params.vcf_location,
'data/rnaGeneFamilyBlacklist.tab', database_map)
# If any of the above were returned as 'DOWNLOAD' then download the radia
# data folder to a temp directory from git and set the values for the
# invalid ones.
if len([db for db, val in database_map.items() if val == 'DOWNLOAD']) > 0:
download_databases(database_map, params.logfile)
# if the -C all option was specified, expand params.chromosome
if params.chromosome == 'all':
params.chromosome = [''.join(['chr', str(i)]) for i in \
range(1, 23)+['X', 'Y']]
return database_map
def process_parameters(params):
'''
This module conducts the error handling for all parmeters passed to the
program.
'''
# Does the provided radia binary provided exist?
params.radia_executable = pi_errors.test_param_value(
params.radia_executable, 'radia', '--radia', params.logfile)
# Setup filterRadia.py
params.filter_radia_executable = '/'.join(
[os.path.split(params.radia_executable)[0], 'filterRadia.py'])
params.filter_radia_executable = pi_errors.test_param_value(
params.filter_radia_executable, 'filterradia', '--radia',
params.logfile)
# Test input files
params.tum_d_file = pi_errors.test_param_value(params.tum_d_file,
'Tumor DNA',
'--tum_dna_file',
params.logfile)
params.norm_d_file = pi_errors.test_param_value(params.norm_d_file,
'Normal DNA',
'--norm_dna_file',
params.logfile)
if params.tum_r_file is not None:
params.tum_r_file = pi_errors.test_param_value(params.tum_r_file,
'Tumor RNA',
'--tum_rna_file',
params.logfile)
# If you don't have a reference, you need twoBitToFasta
if params.index_location is None:
params.tbtf_executable = pi_errors.test_param_value(
params.tbtf_executable, 'twoBitToFa', '--twoBitToFa',
params.logfile)
# Are dnsnp or cosmic vcf required?
if params.dbsnp_file == 'DOWNLOAD' or params.cosmic_file == 'DOWNLOAD' or \
params.genome_fasta == 'DOWNLOAD':
# Ensure the vcf storage location has been provided
if params.vcf_location is None:
raise pi_errors.ParameterError(
dt.now().strftime('%I:%M %p %b %d, %Y') + \
': --vcf_location cannot be empty if either --cosmic, ' + \
'--dbsnp, or --genome_fasta are empty.', params.logfile)
else:
params.vcf_location = os.path.abspath(params.vcf_location)
# Download dbsnp file if required
if params.dbsnp_file == 'DOWNLOAD':
if os.path.exists('/'.join([params.vcf_location, '00-All.vcf'])):
params.dbsnp_file = '/'.join(
[params.vcf_location, '00-All.vcf'])
else:
params.dbsnp_file = prepare.download_vcf('dbsnp', params)
# Download cosmic file if required
if params.cosmic_file == 'DOWNLOAD':
if os.path.exists('/'.join(
[params.vcf_location, 'Cosmic_sorted.vcf'])):
params.cosmic_file = '/'.join(
[params.vcf_location, 'Cosmic_sorted.vcf'])
else:
params.cosmic_file = prepare.download_vcf('cosmic', params)
# Download genome fasta if required
if params.genome_fasta == 'DOWNLOAD' or not \
os.path.exists(params.genome_fasta):
if os.path.exists(''.join(
[params.vcf_location, '/', params.genome_version, '.fa'])):
params.genome_fasta = ''.join(
[params.vcf_location, '/', params.genome_version, '.fa'])
else:
params.genome_fasta = prepare.get_genome(
params.genome_version, params.vcf_location,
params.twoBitToFa_executable, sys.stderr)
else:
params.genome_fasta = os.path.abspath(params.genome_fasta)
# Set up the value for rna_fasta
if params.rna_fasta == 'GENOME_FASTA':
params.rna_fasta = params.genome_fasta
else:
params.rna_fasta = pi_errors.test_param_value(params.rna_fasta,
'RNA Fasta',
'--rna_fasta',
params.logfile)
# Ensure the other databases are set up correctly
# The package path is 2 levels above the
radia_pkg_path = os.path.split(os.path.split(
params.radia_executable)[0])[0]
database_map = defaultdict()
test_database(params.blacklist, 'blacklist', radia_pkg_path,
params.vcf_location, 'data/hg19/blacklists/1000Genomes/' + \
'phase1/', database_map)
test_database(params.retrogenes, 'retrogenes', radia_pkg_path,
params.vcf_location, 'data/hg19/retroGenes/', database_map)
test_database(params.pseudogenes, 'pseudogenes', radia_pkg_path,
params.vcf_location, 'data/hg19/peudoGenes/', database_map)
test_database(params.broad_targets, 'broad_targets', radia_pkg_path,
params.vcf_location, 'data/hg19/broadTargets/', database_map)
test_database(params.rna_blacklist, 'rna_blacklist', radia_pkg_path,
params.vcf_location, 'data/rnaGeneBlacklist.tab',
database_map)
test_database(params.rna_family_blacklist, 'rna_family_blacklist',
radia_pkg_path, params.vcf_location,
'data/rnaGeneFamilyBlacklist.tab', database_map)
# If any of the above were returned as 'DOWNLOAD' then download the radia
# data folder to a temp directory from git and set the values for the
# invalid ones.
if len([db for db, val in database_map.items() if val == 'DOWNLOAD']) > 0:
download_databases(database_map, params.logfile)
# if the -C all option was specified, expand params.chromosome
if params.chromosome == 'all':
params.chromosome = [''.join(['chr', str(i)]) for i in \
range(1, 23)+['X', 'Y']]
return database_map