def test_duplicate_position(self):
reference_position = Position(call='G', coverage='-', proportion='-')
dups_position = Position(call='1', coverage='-', proportion='-')
samples = ((Position(call='G', coverage='-', proportion='-'), ), )
self.analysis.analyze_position(reference_position, dups_position,
samples)
def test_varscan_call_cannot_be_made(self):
"""
VarScan may include a position with ALT values when a call cannot be made.
It should still be called missing (X).
TODO: Add See Also VarScan documentation
"""
# The following is from a SRR011186 sample using bwamem and varscan.
# The positions from the source data were 34072-34074.
vcf_data = (
"#CHROM POS ID REF ALT QUAL FILTER INFO FORMAT SRR011186\n"
"gi|561108321|ref|NC_018143.2| 1 . GC C . PASS ADP=114;WT=0;HET=0;HOM=1;NC=0 GT:GQ:SDP:DP:RD:AD:FREQ:PVAL:RBQ:ABQ:RDF:RDR:ADF:ADR 1/1:255:114:114:6:108:94.74%:1.6043E-58:40:38:2:4:46:62\n"
# This position should be called missing because the GT column is './.'
"gi|561108321|ref|NC_018143.2| 2 . C G . PASS ADP=108;WT=1;HET=0;HOM=0;NC=0 GT:GQ:SDP:DP:RD:AD:FREQ:PVAL:RBQ:ABQ:RDF:RDR:ADF:ADR ./.:.:108\n"
"gi|561108321|ref|NC_018143.2| 3 . A . . PASS ADP=112;WT=1;HET=0;HOM=0;NC=0 GT:GQ:SDP:DP:RD:AD:FREQ:PVAL:RBQ:ABQ:RDF:RDR:ADF:ADR 0/0:209:112:112:111:0:0%:1E0:38:0:47:64:0:0\n"
)
expected = (Position(call='G',
simple_call='G',
coverage=114,
proportion=0.05263157894736842),
Position(call='X',
simple_call='N',
coverage=108.0,
proportion='-'),
Position(call='A',
simple_call='A',
coverage=112,
proportion=0.9910714285714286))
with tempfile.NamedTemporaryFile('w+') as tmpfile:
# Seed the file with test data
tmpfile.write(vcf_data)
tmpfile.seek(0)
# Find the test contig.
vcf = Vcf(tmpfile.name, 'SRR011186', 'varscan', 'bwamem')
contig = vcf.get_contig('gi|561108321|ref|NC_018143.2|')
positions = contig.positions
self.assertIsInstance(contig, VcfContig)
# Check position values.
position = 0
for expect, observe in zip(expected, positions):
position += 1
self.assertEqual(expect, observe)
# It yields all expected positions
self.assertEqual(position, len(expected))
# All following positions should be empty
self.assertEqual(VcfContig.VCF_EMPTY_POSITION, next(positions))
def test_sample_vcfs_return_infinite_positions(self):
vcf = Vcf(testdata.GATK_VCF, 'test_name', 'test_aliner',
'test_snpcaller')
contig = vcf.get_contig('500WT1_test')
positions = contig.positions
expected = (
Position(call='C', simple_call='C', coverage=19049,
proportion='-'),
Position(call='C', simple_call='C', coverage=19049,
proportion='-'),
Position(call='T', simple_call='T', coverage=18824,
proportion='-'),
Position(call='G', simple_call='G', coverage=18804,
proportion='-'),
Position(call='X', simple_call='N', coverage='?', proportion='?'),
Position(call='X', simple_call='N', coverage='?', proportion='?'),
Position(call='G', simple_call='G', coverage=18895,
proportion='-'),
Position(call='A', simple_call='A', coverage=19005,
proportion='-'),
)
# It should yield all the contig positions.
position = 0
for expect, observe in zip(expected, positions):
position += 1
self.assertEqual(expect, observe)
self.assertEqual(len(expected), position)
# It should yield empty positions after the contig is exhausted.
self.assertEqual(VcfContig.VCF_EMPTY_POSITION, next(positions))
self.assertEqual(VcfContig.VCF_EMPTY_POSITION, next(positions))
def test_call_normalizes_to_uppercase(self):
observed = ['g', 'a', 't', 'c', 'd', 'x', '.']
expected = ['G', 'A', 'T', 'C', 'D', 'X', '.']
for observe, expect in zip(observed, expected):
pos = Position(call=observe, coverage='-', proportion='-')
self.assertEqual(expect, pos.call)
def test_simple_call_normalizes_to_uppercase_and_masks_degeneracies_with_N(
self):
observed = ['g', 'a', 't', 'c', 'd', 'x', '.']
expected = ['G', 'A', 'T', 'C', 'N', 'N', 'N']
for observe, expect in zip(observed, expected):
pos = Position(call=observe, coverage='-', proportion='-')
self.assertEqual(expect, pos.simple_call)
def test_unique_position(self):
reference_position = Position(call='G', coverage='-', proportion='-')
dups_position = Position(
# Unique position
call='0',
coverage='-',
proportion='-')
samples = (
(
# Single Nucleotide Monomorphism
Position(call='G', coverage='-', proportion='-'),
# Single Nucleotide Polymorphism
Position(call='A', coverage='-', proportion='-'),
), )
self.analysis.analyze_position(reference_position, dups_position,
samples)
def test_no_duplicate_information_position(self):
"""
Scenario: The reference was not scanned for duplicate positions.
As a result, it is assumed all positions passed.
The following fields should be affected:
- is_reference_duplicated is False
- called_snp is incremented for snps, but not for monomorphisms
- is_missing_matrix is True
- is_all_quality_breadth is True
"""
reference_position = Position(call='G', coverage='-', proportion='-')
dups_position = Position(
# No duplicate information
call='X',
coverage='-',
proportion='-')
samples = (
(
# Single Nucleotide Monomorphism
Position(call='G', coverage='-', proportion='-'),
# Single Nucleotide Polymorphism
Position(call='A', coverage='-', proportion='-'),
), )
expected = PositionInfo(is_all_called=True,
is_reference_clean=True,
is_reference_duplicated=False,
is_all_passed_coverage=True,
is_all_passed_proportion=True,
is_all_passed_consensus=True,
is_all_quality_breadth=True,
is_best_snp=False,
all_sample_stats=[
(Counter({
'quality_breadth': 1,
'was_called': 1,
'called_reference': 1,
'passed_proportion_filter': 1,
'passed_coverage_filter': 1,
'called_snp': 0,
'called_degen': 0
}),
Counter({
'quality_breadth': 1,
'was_called': 1,
'called_reference': 1,
'passed_proportion_filter': 1,
'passed_coverage_filter': 1,
'called_snp': 0,
'called_degen': 0
})),
[
Counter({
'quality_breadth': 1,
'was_called': 1,
'called_reference': 1,
'passed_proportion_filter': 1,
'passed_coverage_filter': 1,
'called_snp': 0,
'called_degen': 0
}),
Counter({
'quality_breadth': 1,
'was_called': 1,
'called_reference': 1,
'passed_proportion_filter': 1,
'passed_coverage_filter': 1,
'called_snp': 0,
'called_degen': 0
}),
Counter({
'quality_breadth': 1,
'was_called': 1,
'called_reference': 1,
'passed_proportion_filter': 1,
'passed_coverage_filter': 1,
'called_snp': 0,
'called_degen': 0
})
]
],
is_missing_matrix=False,
called_reference=1,
called_snp=0,
passed_coverage_filter=1,
passed_proportion_filter=1,
num_A=0,
num_C=0,
num_G=1,
num_T=0,
num_N=0,
call_str=['G', 'G'],
masked_call_str=['G', 'G'],
CallWasMade='Y',
PassedDepthFilter='-',
PassedProportionFilter='-',
Pattern=['1', '1'])
expected = PositionInfo(
is_all_called=True,
is_reference_clean=True,
is_reference_duplicated=False,
is_all_passed_coverage=True,
is_all_passed_proportion=True,
is_all_passed_consensus=False,
is_all_quality_breadth=False,
is_best_snp=False,
all_sample_stats=[[
Counter({
'quality_breadth': 1,
'called_reference': 1,
'called_snp': 1,
'passed_coverage_filter': 1,
'passed_proportion_filter': 1,
'was_called': 1,
'called_degen': 0
}),
Counter({
'quality_breadth': 1,
'passed_coverage_filter': 1,
'passed_proportion_filter': 1,
'was_called': 1,
'called_reference': 0,
'called_snp': 0,
'called_degen': 0
})
],
[
Counter({
'quality_breadth': 1,
'called_reference': 1,
'called_snp': 1,
'passed_coverage_filter': 1,
'passed_proportion_filter': 1,
'was_called': 1,
'called_degen': 0
}),
Counter({
'quality_breadth': 1,
'passed_coverage_filter': 1,
'passed_proportion_filter': 1,
'was_called': 1,
'called_reference': 0,
'called_snp': 0,
'called_degen': 0
}),
Counter({
'quality_breadth': 1,
'called_reference': 1,
'passed_coverage_filter': 1,
'passed_proportion_filter': 1,
'was_called': 1,
'called_snp': 0,
'called_degen': 0
}),
Counter({
'quality_breadth': 1,
'called_snp': 1,
'passed_coverage_filter': 1,
'passed_proportion_filter': 1,
'was_called': 1,
'called_reference': 0,
'called_degen': 0
})
]],
is_missing_matrix=True,
called_reference=1,
called_snp=1,
passed_coverage_filter=2,
passed_proportion_filter=2,
num_A=1,
num_C=0,
num_G=1,
num_T=0,
num_N=0,
call_str=['G', 'G', 'A'],
masked_call_str=['G', 'G', 'A'],
CallWasMade='YY',
PassedDepthFilter='--',
PassedProportionFilter='--',
Pattern=['1', '1', '2'])
self.assertEqual(
expected,
self.analysis.analyze_position(reference_position, dups_position,
samples))
def test_fasta_position(self):
reference_position = Position(call='G', coverage='-', proportion='-')
dups_position = Position(call='0', coverage='-', proportion='-')
samples = ((Position(call='G', coverage='-', proportion='-'),
Position(call='G', coverage='-', proportion='-')), )
expected = PositionInfo(
is_all_called=True,
is_reference_clean=True,
is_reference_duplicated=False,
is_all_passed_coverage=True,
is_all_passed_proportion=True,
is_all_passed_consensus=True,
is_all_quality_breadth=True,
is_best_snp=False,
all_sample_stats=[[
Counter({
'called_reference': 1,
'passed_coverage_filter': 1,
'quality_breadth': 1,
'was_called': 1,
'passed_proportion_filter': 1,
'called_snp': 0,
'called_degen': 0
}),
Counter({
'called_reference': 1,
'passed_coverage_filter': 1,
'quality_breadth': 1,
'was_called': 1,
'passed_proportion_filter': 1,
'called_snp': 0,
'called_degen': 0
})
],
[
Counter({
'called_reference': 1,
'passed_coverage_filter': 1,
'quality_breadth': 1,
'was_called': 1,
'passed_proportion_filter': 1,
'called_snp': 0,
'called_degen': 0
}),
Counter({
'called_reference': 1,
'passed_coverage_filter': 1,
'quality_breadth': 1,
'was_called': 1,
'passed_proportion_filter': 1,
'called_snp': 0,
'called_degen': 0
}),
Counter({
'called_reference': 1,
'passed_coverage_filter': 1,
'quality_breadth': 1,
'was_called': 1,
'passed_proportion_filter': 1,
'called_snp': 0,
'called_degen': 0
}),
Counter({
'called_reference': 1,
'passed_coverage_filter': 1,
'quality_breadth': 1,
'was_called': 1,
'passed_proportion_filter': 1,
'called_snp': 0,
'called_degen': 0
})
]],
is_missing_matrix=False,
called_reference=2,
called_snp=0,
passed_coverage_filter=2,
passed_proportion_filter=2,
num_A=0,
num_C=0,
num_G=2,
num_T=0,
num_N=0,
call_str=['G', 'G', 'G'],
masked_call_str=['G', 'G', 'G'],
CallWasMade='YY',
PassedDepthFilter='--',
PassedProportionFilter='--',
Pattern=['1', '1', '1'])
self.assertEqual(
expected,
self.analysis.analyze_position(reference_position, dups_position,
samples))
def test_duplicate_information_position(self):
"""
Scenario: The reference was scanned for duplicate positions.
Positions marked as duplicates are not quality positions and should not increment the sample statistics.
- is_reference_duplicated is True
- is_missing_matrix is False
- is_all_quality_breadth is False
- is_all_passed_consensus is False
- None of the sample stats are incremented
"""
reference_position = Position(call='G', coverage='-', proportion='-')
dups_position = Position(
# No duplicate information
call='1',
coverage='-',
proportion='-')
samples = (
(
# Single Nucleotide Monomorphism
Position(call='G', coverage='-', proportion='-'),
# Single Nucleotide Polymorphism
Position(call='A', coverage='-', proportion='-'),
), )
expected = PositionInfo(
is_all_called=True,
is_reference_clean=True,
is_reference_duplicated=True,
is_all_passed_coverage=True,
is_all_passed_proportion=True,
is_all_passed_consensus=False,
is_all_quality_breadth=False,
is_best_snp=False,
all_sample_stats=[[
Counter({
'passed_coverage_filter': 1,
'was_called': 1,
'passed_proportion_filter': 1,
'called_reference': 0,
'quality_breadth': 0,
'called_snp': 0,
'called_degen': 0
}),
Counter({
'passed_coverage_filter': 1,
'was_called': 1,
'passed_proportion_filter': 1,
'called_reference': 0,
'quality_breadth': 0,
'called_snp': 0,
'called_degen': 0
})
],
[
Counter({
'passed_coverage_filter': 1,
'was_called': 1,
'passed_proportion_filter': 1,
'called_reference': 0,
'quality_breadth': 0,
'called_snp': 0,
'called_degen': 0
}),
Counter({
'passed_coverage_filter': 1,
'was_called': 1,
'passed_proportion_filter': 1,
'called_reference': 0,
'quality_breadth': 0,
'called_snp': 0,
'called_degen': 0
}),
Counter({
'passed_coverage_filter': 1,
'was_called': 1,
'passed_proportion_filter': 1,
'called_reference': 0,
'quality_breadth': 0,
'called_snp': 0,
'called_degen': 0
}),
Counter({
'passed_coverage_filter': 1,
'was_called': 1,
'passed_proportion_filter': 1,
'called_reference': 0,
'quality_breadth': 0,
'called_snp': 0,
'called_degen': 0
})
]],
is_missing_matrix=False,
called_reference=1,
called_snp=0,
passed_coverage_filter=2,
passed_proportion_filter=2,
num_A=1,
num_C=0,
num_G=1,
num_T=0,
num_N=0,
call_str=['G', 'G', 'A'],
masked_call_str=['G', 'G', 'A'],
CallWasMade='YY',
PassedDepthFilter='--',
PassedProportionFilter='--',
Pattern=['1', '1', '2'])
self.assertEqual(
expected,
self.analysis.analyze_position(reference_position, dups_position,
samples))
def setUp(self):
# Source fasta values
contigs = (
# No gap
(">contig0\n"
"GATC\n"
"GGAA\n"),
# Gap after contig
(
">contig1\n"
"GATC\n"
"GGAA\n"
"\n"
# ),
# # No linebreak
# (
# ">contig2\n"
# "GATCGGAA"
)
# TODO: >80 characters contig?
)
# Expected values
self.contigs_expected = ({
'name':
'contig0',
'file_position':
0 + len('>contig0\n'),
'positions': (Position(call='G', coverage='-', proportion='-'),
Position(call='A', coverage='-', proportion='-'),
Position(call='T', coverage='-', proportion='-'),
Position(call='C', coverage='-', proportion='-'),
Position(call='G', coverage='-', proportion='-'),
Position(call='G', coverage='-', proportion='-'),
Position(call='A', coverage='-', proportion='-'),
Position(call='A', coverage='-', proportion='-'))
}, {
'name':
'contig1',
'file_position':
len(contigs[0]) + len('>contig1\n'),
'positions': (Position(call='G', coverage='-', proportion='-'),
Position(call='A', coverage='-', proportion='-'),
Position(call='T', coverage='-', proportion='-'),
Position(call='C', coverage='-', proportion='-'),
Position(call='G', coverage='-', proportion='-'),
Position(call='G', coverage='-', proportion='-'),
Position(call='A', coverage='-', proportion='-'),
Position(call='A', coverage='-', proportion='-'))
})
fasta_content = "".join(contigs)
# Create a mock fasta
self.fasta_file = tempfile.NamedTemporaryFile(mode='w+', delete=False)
self.fasta_file.write(fasta_content)
self.fasta_file.seek(0)
# Instantiate test contigs
self.contig0 = FastaContig(self.contigs_expected[0]['name'],
len(self.contigs_expected[0]['positions']),
self.contigs_expected[0]['file_position'],
self.fasta_file.name,
is_reference=False)
self.contig1 = FastaContig(self.contigs_expected[1]['name'],
len(self.contigs_expected[1]['positions']),
self.contigs_expected[1]['file_position'],
self.fasta_file.name,
is_reference=False)
self.ref_contig0 = FastaContig(
self.contigs_expected[0]['name'],
len(self.contigs_expected[0]['positions']),
self.contigs_expected[0]['file_position'],
self.fasta_file.name,
is_reference=True)
self.ref_contig1 = FastaContig(
self.contigs_expected[1]['name'],
len(self.contigs_expected[1]['positions']),
self.contigs_expected[1]['file_position'],
self.fasta_file.name,
is_reference=True)
def test_multi_base_call_raises_exception(self):
# TODO: use appropriate exception
with self.assertRaises(Exception):
Position(call='gatc', coverage='-', proportion='-')