def test_open_known_genes_missing_lines(self):
''' test that open_known_genes() works correctly when we can't find any genes
'''
header = ['gene', 'chr', 'start', 'stop', 'type', 'mode', 'mech', 'hgnc_id']
self.temp.write(('\t'.join(header) + '\n').encode('utf8'))
self.temp.flush()
# if we have checked the file, and there aren't any genes in it, this
# raises an error, since the most likely explanation is that something
# has gone wrong with the data file, and likely the line-endings
with self.assertRaises(ValueError):
open_known_genes(self.temp.name)
def test_open_known_genes_missing_lines(self):
''' test that open_known_genes() works correctly when we can't find any genes
'''
header = [
'gene', 'chr', 'start', 'stop', 'type', 'mode', 'mech', 'hgnc_id'
]
self.temp.write(('\t'.join(header) + '\n').encode('utf8'))
self.temp.flush()
# if we have checked the file, and there aren't any genes in it, this
# raises an error, since the most likely explanation is that something
# has gone wrong with the data file, and likely the line-endings
with self.assertRaises(ValueError):
open_known_genes(self.temp.name)
def test_open_known_genes_wrong_status(self):
''' test that open_known_genes() filters out genes without a good status
'''
header = [
'gene', 'chr', 'start', 'stop', 'type', 'mode', 'mech', 'hgnc_id'
]
line1 = [
'TEST', '1', '1000', '2000', 'possible dd gene', 'Monoallelic',
'Loss-of-function', '1001'
]
line2 = [
'TEST2', '1', '3000', '4000', 'confirmed dd gene', 'Monoallelic',
'Loss-of-function', '2001'
]
self.temp.write(('\t'.join(header) + '\n').encode('utf8'))
self.temp.write(('\t'.join(line1) + '\n').encode('utf8'))
self.temp.write(('\t'.join(line2) + '\n').encode('utf8'))
self.temp.flush()
self.assertEqual(
open_known_genes(self.temp.name), {
'2001': {
'chrom': '1',
'start': 3000,
'end': 4000,
'symbol': 'TEST2',
'status': set(['confirmed dd gene']),
'inh': {
'Monoallelic': set(['Loss-of-function'])
}
}
})
def test_open_known_genes_multimechs(self):
''' test that open_known_genes() works correctly for genes with >1 mechs
'''
header = [
'gene', 'chr', 'start', 'stop', 'type', 'mode', 'mech', 'hgnc_id'
]
line1 = [
'TEST', '1', '1000', '2000', 'confirmed dd gene', 'Monoallelic',
'Loss-of-function', '1001'
]
line2 = [
'TEST', '1', '1000', '2000', 'confirmed dd gene', 'Monoallelic',
'Activating', '1001'
]
self.temp.write(('\t'.join(header) + '\n').encode('utf8'))
self.temp.write(('\t'.join(line1) + '\n').encode('utf8'))
self.temp.write(('\t'.join(line2) + '\n').encode('utf8'))
self.temp.flush()
self.assertEqual(
open_known_genes(self.temp.name), {
'1001': {
'chrom': '1',
'start': 1000,
'end': 2000,
'symbol': 'TEST',
'status': set(['confirmed dd gene']),
'inh': {
'Monoallelic': set(['Loss-of-function', 'Activating'])
}
}
})
def test_open_known_genes(self):
''' test that open_known_genes() works correctly
'''
header = ['gene', 'chr', 'start', 'stop', 'type', 'mode', 'mech', 'hgnc_id']
line = ['TEST', '1', '1000', '2000', 'confirmed dd gene',
'Monoallelic', 'Loss-of-function', '1001']
self.temp.write(('\t'.join(header) + '\n').encode('utf8'))
self.temp.write(('\t'.join(line) + '\n').encode('utf8'))
self.temp.flush()
self.assertEqual(open_known_genes(self.temp.name),
{'1001': {'chrom': '1', 'start': 1000, 'end': 2000,
'symbol': 'TEST', 'status': set(['confirmed dd gene']),
'inh': {'Monoallelic': set(['Loss-of-function'])}}
})
def __init__(self, population_tags=None, count=0, known_genes=None, date=None,
regions=None, lof_sites=None, pp_filter=0.0, sum_x_lr2_file=None,
output_path=None, export_vcf=None, debug_chrom=None, debug_pos=None):
""" initialise the class object
Args:
population_tags: list of population ID tags, that could exist within
the INFO field, or None.
count: number of probands to analyse, helpful for tracking progress
in output logs.
known_genes: path to table of genes genes known to be associated
with genetic disorders, or None.
date: date of the known_genes file, or None if not using/unknown.
regions: path to a table of regions for DECIPHER CNV syndromes.
lof_sites: path to json file of [chrom, position] coordinates in
genome, for modifying to a loss-of-function consequence if
required. Can be None if unneeded.
pp_filter: threshold from 0 to 1 for pp_dnm value to filter out
candidiate DNMs which fall below this value
sum_x_lr2_file: File containing sum of l2r values on x chromosome
for each person
output_path: path to write output tab-separated file to
export_vcf: path to file or folder to write VCFs to.
debug_chrom: chromosome for debugging purposes.
debug_pos: position for debugging variant filtering at.
"""
self.pp_filter = pp_filter
self.total = count
self.count = 0
self.populations = population_tags
self.debug_chrom = debug_chrom
self.debug_pos = debug_pos
# open reference datasets, these return None if the paths are None
self.known_genes = open_known_genes(known_genes)
self.cnv_regions = open_cnv_regions(regions)
self.last_base = open_last_base_sites(lof_sites)
#open file containing sum of mean log 2 ratios on X, returns an empty dict if path is None
self.sum_x_lr2 = open_x_lr2_file(sum_x_lr2_file)
self.reporter = Report(output_path, export_vcf, date)
def test_open_known_genes_wrong_status(self):
''' test that open_known_genes() filters out genes without a good status
'''
header = ['gene', 'chr', 'start', 'stop', 'type', 'mode', 'mech', 'hgnc_id']
line1 = ['TEST', '1', '1000', '2000', 'possible dd gene',
'Monoallelic', 'Loss-of-function', '1001']
line2 = ['TEST2', '1', '3000', '4000', 'confirmed dd gene',
'Monoallelic', 'Loss-of-function', '2001']
self.temp.write(('\t'.join(header) + '\n').encode('utf8'))
self.temp.write(('\t'.join(line1) + '\n').encode('utf8'))
self.temp.write(('\t'.join(line2) + '\n').encode('utf8'))
self.temp.flush()
self.assertEqual(open_known_genes(self.temp.name),
{'2001': {'chrom': '1', 'start': 3000, 'end': 4000,
'symbol': 'TEST2', 'status': set(['confirmed dd gene']),
'inh': {'Monoallelic': set(['Loss-of-function'])}}
})
def test_open_known_genes_multimechs(self):
''' test that open_known_genes() works correctly for genes with >1 mechs
'''
header = ['gene', 'chr', 'start', 'stop', 'type', 'mode', 'mech']
line1 = ['TEST', '1', '1000', '2000', 'confirmed dd gene',
'Monoallelic', 'Loss-of-function']
line2 = ['TEST', '1', '1000', '2000', 'confirmed dd gene',
'Monoallelic', 'Activating']
self.temp.write(('\t'.join(header) + '\n').encode('utf8'))
self.temp.write(('\t'.join(line1) + '\n').encode('utf8'))
self.temp.write(('\t'.join(line2) + '\n').encode('utf8'))
self.temp.flush()
self.assertEqual(open_known_genes(self.temp.name),
{'TEST': {'chrom': '1', 'start': 1000, 'end': 2000,
'status': set(['confirmed dd gene']),
'inh': {'Monoallelic': set(['Loss-of-function', 'Activating'])}}
})
def load_definitions_files(self):
"""loads all the config files for the script (eg filters, gene IDs)
"""
# if we have named a gene file, then load a dictionary of genes, and
# add them to the filters, so we can screen variants for being in genes
# known to be involved with disorders
self.known_genes = None
if self.options.genes is not None:
self.known_genes = open_known_genes(self.options.genes)
# if we have named an ID mapping file, the load a dictionary of IDs and
# alternate IDs, so we can convert between different ID schemes.
self.ID_mapper = None
if self.options.alternate_ids is not None:
self.ID_mapper = create_person_ID_mapper(self.options.alternate_ids)
# open a list of regions associated with DECIPHER syndromes
self.cnv_regions = None
if self.options.regions is not None:
self.cnv_regions = open_cnv_regions(self.options.regions)
def load_definitions_files(self):
"""loads all the config files for the script (eg filters, gene IDs)
"""
# if we have named a gene file, then load a dictionary of genes, and
# add them to the filters, so we can screen variants for being in genes
# known to be involved with disorders
self.known_genes = None
if self.options.genes is not None:
self.known_genes = open_known_genes(self.options.genes)
# if we have named an ID mapping file, the load a dictionary of IDs and
# alternate IDs, so we can convert between different ID schemes.
self.ID_mapper = None
if self.options.alternate_ids is not None:
self.ID_mapper = create_person_ID_mapper(
self.options.alternate_ids)
# open a list of regions associated with DECIPHER syndromes
self.cnv_regions = None
if self.options.regions is not None:
self.cnv_regions = open_cnv_regions(self.options.regions)
