def getOptionalArgs():
parser = argparse.ArgumentParser(add_help=False)
optional = parser.add_argument_group('Optional arguments')
optional.add_argument("--help",
"-h",
action="help",
help="show this help message and exit")
optional.add_argument(
'--scaleFactorsMethod',
help='Method to use to scale the samples. '
'If a method is specified, then it will be used to compensate '
'for sequencing depth differences between the samples. '
'As an alternative, this can be set to None and an option from '
'--normalizeUsing <method> can be used. (Default: %(default)s)',
choices=['readCount', 'SES', 'None'],
default='readCount')
optional.add_argument(
'--sampleLength',
'-l',
help='*Only relevant when SES is chosen for the '
'scaleFactorsMethod.* To compute the SES, specify '
'the length (in bases) of the regions (see --numberOfSamples) '
'that will be randomly sampled to calculate the scaling factors. '
'If you do not have a good sequencing depth for '
'your samples consider increasing the sampling '
'regions\' size to minimize the probability '
'that zero-coverage regions are used. (Default: %(default)s)',
default=1000,
type=int)
optional.add_argument(
'--numberOfSamples',
'-n',
help='*Only relevant when SES is chosen for the '
'scaleFactorsMethod.* Number of samplings taken '
'from the genome to compute the scaling factors. (Default: %(default)s)',
default=1e5,
type=int)
optional.add_argument(
'--scaleFactors',
help='Set this parameter manually to avoid the computation of '
'scaleFactors. The format is scaleFactor1:scaleFactor2.'
'For example, --scaleFactor 0.7:1 will cause the first BAM file to'
'be multiplied by 0.7, while not scaling '
'the second BAM file (multiplication with 1).',
default=None,
required=False)
optional.add_argument(
'--operation',
help='The default is to output the log2 ratio of the '
'two samples. The reciprocal ratio returns the '
'the negative of the inverse of the ratio '
'if the ratio is less than 0. The resulting '
'values are interpreted as negative fold changes. '
'Instead of performing a computation using both files, the scaled signal can '
'alternatively be output for the first or second file using '
'the \'--operation first\' or \'--operation second\'. (Default: %(default)s)',
default='log2',
choices=[
'log2', 'ratio', 'subtract', 'add', 'mean', 'reciprocal_ratio',
'first', 'second'
],
required=False)
optional.add_argument(
'--pseudocount',
help='A small number to avoid x/0. Only useful '
'together with --operation log2 or --operation ratio. '
'You can specify different values as pseudocounts for '
'the numerator and the denominator by providing two '
'values (the first value is used as the numerator '
'pseudocount and the second the denominator pseudocount). (Default: %(default)s)',
default=[1],
type=float,
nargs='+',
action=parserCommon.requiredLength(1, 2),
required=False)
optional.add_argument(
'--skipZeroOverZero',
help='Skip bins where BOTH BAM files lack coverage. '
'This is determined BEFORE any applicable pseudocount '
'is added.',
action='store_true')
return parser
def parse_arguments(args=None):
parentParser = parserCommon.getParentArgParse()
outputParser = parserCommon.output()
dbParser = parserCommon.deepBlueOptionalArgs()
parser = argparse.ArgumentParser(
parents=[parentParser, outputParser, dbParser],
formatter_class=argparse.ArgumentDefaultsHelpFormatter,
description='This tool compares two bigWig files based on the number '
'of mapped reads. To compare the bigWig files, the genome is '
'partitioned into bins of equal size, then the number of reads found '
'in each BAM file are counted per bin and finally a summary '
'value is reported. This value can be the ratio of the number of reads'
'per bin, the log2 of the ratio, the sum or the difference.')
# define the arguments
parser.add_argument('--bigwig1',
'-b1',
metavar='Bigwig file',
help='Bigwig file 1. Usually the file for the '
'treatment.',
required=True)
parser.add_argument('--bigwig2',
'-b2',
metavar='Bigwig file',
help='Bigwig file 2. Usually the file for the '
'control.',
required=True)
parser.add_argument('--scaleFactors',
help='Set this parameter to multipy the bigwig values '
'by a constant. The format is '
'scaleFactor1:scaleFactor2. '
'For example 0.7:1 to scale the first bigwig file '
'by 0.7 while not scaling the second bigwig file',
default=None,
required=False)
parser.add_argument(
'--pseudocount',
help='A small number to avoid x/0. Only useful '
'together with --operation log2 or --operation ratio. '
'You can specify different values as pseudocounts for '
'the numerator and the denominator by providing two '
'values (the first value is used as the numerator '
'pseudocount and the second the denominator pseudocount). (Default: %(default)s)',
default=1,
nargs='+',
action=parserCommon.requiredLength(1, 2),
type=float,
required=False)
parser.add_argument('--skipZeroOverZero',
help='Skip bins where BOTH BAM files lack coverage. '
'This is determined BEFORE any applicable pseudocount '
'is added.',
action='store_true')
parser.add_argument(
'--operation',
help='The default is to output the log2ratio of the '
'two samples. The reciprocal ratio returns the '
'the negative of the inverse of the ratio '
'if the ratio is less than 0. The resulting '
'values are interpreted as negative fold changes. '
'Instead of performing a '
'computation using both files, the scaled signal can '
'alternatively be output for the first or second file using '
'the \'--operation first\' or \'--operation second\' (Default: %(default)s)',
default='log2',
choices=[
'log2', 'ratio', 'subtract', 'add', 'mean', 'reciprocal_ratio',
'first', 'second'
],
required=False)
parser.add_argument(
'--skipNonCoveredRegions',
'--skipNAs',
help=
'This parameter determines if non-covered regions (regions without a score) '
'in the bigWig files should be skipped. The default is to treat those '
'regions as having a value of zero. '
'The decision to skip non-covered regions '
'depends on the interpretation of the data. Non-covered regions '
'in a bigWig file may represent repetitive regions that should '
'be skipped. Alternatively, the interpretation of non-covered regions as '
'zeros may be wrong and this option should be used ',
action='store_true')
return parser
def getOptionalArgs():
parser = argparse.ArgumentParser(add_help=False)
optional = parser.add_argument_group('Optional arguments')
optional.add_argument("--help", "-h", action="help",
help="show this help message and exit")
optional.add_argument('--scaleFactorsMethod',
help='Method to use to scale the samples. '
'If a method is specified, then it will be used to compensate '
'for sequencing depth differences between the samples. '
'As an alternative, this can be set to None and an option from '
'--normalizeUsing <method> can be used.',
choices=['readCount', 'SES', 'None'],
default='readCount')
optional.add_argument('--sampleLength', '-l',
help='*Only relevant when SES is chosen for the '
'scaleFactorsMethod.* To compute the SES, specify '
'the length (in bases) of the regions (see --numberOfSamples) '
'that will be randomly sampled to calculate the scaling factors. '
'If you do not have a good sequencing depth for '
'your samples consider increasing the sampling '
'regions\' size to minimize the probability '
'that zero-coverage regions are used.',
default=1000,
type=int)
optional.add_argument('--numberOfSamples', '-n',
help='*Only relevant when SES is chosen for the '
'scaleFactorsMethod.* Number of samplings taken '
'from the genome to compute the scaling factors.',
default=1e5,
type=int)
optional.add_argument('--scaleFactors',
help='Set this parameter manually to avoid the computation of '
'scaleFactors. The format is scaleFactor1:scaleFactor2.'
'For example, --scaleFactor 0.7:1 will cause the first BAM file to'
'be multiplied by 0.7, while not scaling '
'the second BAM file (multiplication with 1).',
default=None,
required=False)
optional.add_argument('--operation',
help='The default is to output the log2 ratio of the '
'two samples. The reciprocal ratio returns the '
'the negative of the inverse of the ratio '
'if the ratio is less than 0. The resulting '
'values are interpreted as negative fold changes. '
'Instead of performing a computation using both files, the scaled signal can '
'alternatively be output for the first or second file using '
'the \'--operation first\' or \'--operation second\'',
default='log2',
choices=['log2', 'ratio', 'subtract', 'add', 'mean',
'reciprocal_ratio', 'first', 'second'],
required=False)
optional.add_argument('--pseudocount',
help='A small number to avoid x/0. Only useful '
'together with --operation log2 or --operation ratio. '
'You can specify different values as pseudocounts for '
'the numerator and the denominator by providing two '
'values (the first value is used as the numerator '
'pseudocount and the second the denominator pseudocount).',
default=[1],
type=float,
nargs='+',
action=parserCommon.requiredLength(1, 2),
required=False)
optional.add_argument('--skipZeroOverZero',
help='Skip bins where BOTH BAM files lack coverage. '
'This is determined BEFORE any applicable pseudocount '
'is added.',
action='store_true')
return parser