def main(args=None):
"""
The main method for aliquot-maf-tools.
"""
# Setup logger
Logger.setup_root_logger()
logger = Logger.get_logger("main")
# Print header
logger.info('-'*75)
logger.info("Program Args: aliquot-maf-tools " + " ".join(sys.argv[1::]))
logger.info('Date/time: {0}'.format(datetime.datetime.now()))
logger.info('-'*75)
logger.info('-'*75)
# Get args
p = argparse.ArgumentParser("GDC Aliquot MAF Tools")
subparsers = p.add_subparsers(dest="subcommand")
subparsers.required = True
VcfToAliquotMaf.add(subparsers=subparsers)
MergeAliquotMafs.add(subparsers=subparsers)
MaskMergedAliquotMaf.add(subparsers=subparsers)
options = p.parse_args(args)
# Run
cls = options.func(options)
cls.do_work()
# Finish
logger.info("Finished!")
def __init__(self, options=dict()):
self.logger = Logger.get_logger(self.__class__.__name__)
self.options = options
# Maf stuff
self.maf_header = None
self.maf_writer = None
self._scheme = None
self._columns = None
self._colset = None
def __init__(self, options=dict()):
self.logger = Logger.get_logger(self.__class__.__name__)
self.options = options
self.maf_readers = []
self.callers = []
self.maf_writer = None
self._scheme = None
self._columns = None
self._colset = None
self._merger = None
def __init__(self, options=dict()):
self.logger = Logger.get_logger(self.__class__.__name__)
self.options = options
self.maf_reader = None
self.maf_writer = None
self._scheme = None
self._columns = None
self._colset = None
self.metrics = MafMetricsCollection()
def __init__(self, scheme):
"""
Initialize the MAF record merging object which has the main `merge_records`
function to take an `aliquotmaf.merging.overlap_set.OverlapSet`
instance and performs merging.
"""
self.logger = Logger.get_logger(self.__class__.__name__)
self.scheme = scheme
self.columns = scheme.column_names()
self.logger.info("Loading MAF record merger...")
def __init__(self, name=None, source=None, scheme=None):
self.name = None
self.source = source
self.scheme = scheme
self.logger = Logger.get_logger(self.__class__.__name__)
class GenotypeAndDepthsExtractor(Extractor):
"""Extractor class for extracting the genotype and depths based on the
variant allele index.
"""
logger = Logger.get_logger('GenotypeAndDepthsExtractor')
@classmethod
def extract(cls, var_allele_idx, genotype, alleles):
"""
Extracts the information for the variant alleles based on the
variant allele index. Creates a new, updated genotype record
and depths list.
:param var_allele_idx: the variant allele index
:param genotype: a dictionary or dictionary-like object containing
various possible keys like AD, DP, etc.
:param alleles: an ordered list or tuple of the possible alleles
at the locus
:returns: an updated genotype record and depths list
"""
depths = []
new_gt = {}
if not genotype['GT']: return new_gt, depths
# If DP is defined, set it in new_gt
if 'DP' in genotype and genotype['DP'] is not None:
new_gt['DP'] = genotype['DP']
# If AD is defined, then parse out all REF/ALT allele depths, or whatever is in it
if 'AD' in genotype and genotype['AD'] is not None:
if isinstance(genotype['AD'], int):
depths = [genotype['AD']]
else:
depths = list(genotype['AD'])
# handle VarScan VCF lines where AD contains only 1 depth, and REF allele depth is in RD
if len(depths) == 1 and 'RD' in genotype:
depths = [None for i in alleles]
depths[0] = genotype['RD']
if isinstance(genotype['AD'], int):
depths[var_allele_idx] = genotype['AD']
else:
depths[var_allele_idx] = genotype['AD'][0]
# Handle SomaticSniper VCF lines, where allele depths must be extracted from BCOUNT
elif 'AD' not in genotype and 'BCOUNT' in genotype:
b_idx = {'A': 0, 'C': 1, 'G': 2, 'T': 3}
bcount = list(genotype['BCOUNT'])
depths = [
bcount[b_idx[i]] if i in b_idx else None for i in alleles
]
# If N depths not equal to N alleles, blank out the depths
elif depths and len(depths) != len(alleles):
cls.logger.warning(
'The length of DP array != length of allele array')
depths = [None for i in alleles]
# Sanity check that REF/ALT allele depths are lower than total depth
if 'DP' in genotype and genotype['DP'] is not None and ( \
(depths[0] is not None and depths[0] > genotype['DP']) or \
(depths[var_allele_idx] is not None and depths[var_allele_idx] > genotype['DP']) or \
(depths[0] is not None and depths[var_allele_idx] is not None \
and depths[0] + depths[var_allele_idx] > genotype['DP'])):
cls.logger.warning(
'REF/ALT allele depths are lower than total depth!!')
new_gt['DP'] = 0
for i in depths:
if i and i != '.':
new_gt['DP'] += i
## If depths is empty, just set to 0, 0
if not depths: depths = [0, 0]
## If we have REF/ALT allele depths but not DP, then set DP equal to sum of all ADs
if ((depths[0] is not None and depths[var_allele_idx] is not None) \
and ('DP' not in genotype or genotype['DP'] is None or genotype['DP'] == '.')):
#cls.logger.warn('Missing DP field. setting DP equal to sum of ADs!!')
new_gt['DP'] = sum([i for i in depths if i and i != '.'])
## Set the formatted AD and alleles
new_gt['AD'] = tuple(
[i if i != "" and i is not None else "." for i in depths])
new_gt['GT'] = genotype['GT']
depths = [i if i != "." and i is not None else 0 for i in new_gt['AD']]
return new_gt, depths