def test_iterate_exclude_filters(self):
reads = Reads(self.sam1_file)
reads.filter_unmapped(queue=True)
reads.filter_quality(35, queue=True)
reads.filter_non_unique(strict=True, queue=True)
reads.run_queued_filters()
assert len(
list(
reads.reads(
excluded_filters=['unmapped', 'mapq', 'uniqueness'
]))) == 271
assert len(
list(reads.reads(
excluded_filters=['unmapped', 'uniqueness']))) == 246
assert len(list(
reads.reads(excluded_filters=['unmapped', 'mapq']))) == 153
assert len(list(
reads.reads(excluded_filters=['mapq', 'uniqueness']))) == 271
assert len(list(reads.reads(excluded_filters=['unmapped']))) == 153
assert len(list(reads.reads(excluded_filters=['mapq']))) == 153
assert len(list(reads.reads(excluded_filters=['uniqueness']))) == 246
reads.close()
class TestAccessOptimisedReadPairs(TestFragmentMappedReadPairs):
@classmethod
def setup_method(self, method):
self.dir = os.path.dirname(os.path.realpath(__file__))
sam1_file = self.dir + "/test_seq/lambda_reads1.sam"
sam2_file = self.dir + "/test_seq/lambda_reads2.sam"
self.reads1 = Reads(sam1_file)
self.reads2 = Reads(sam2_file)
self.reads1.filter_unmapped()
self.reads2.filter_unmapped()
self.genome = Genome.from_folder(self.dir + "/test_seq/lambda_genome/")
self.pairs_class = AccessOptimisedReadPairs
self.pairs = self.pairs_class()
regions = self.genome.get_regions(1000)
self.pairs.load(self.reads1, self.reads2, regions=regions)
regions.close()
def teardown_method(self, method):
self.reads1.close()
self.reads2.close()
self.genome.close()
self.pairs.close()
class TestBWAFragmentMappedReads:
def setup_method(self, method):
self.dir = os.path.dirname(os.path.realpath(__file__))
sam1_file = self.dir + "/test_seq/test_bwa1.sam"
sam2_file = self.dir + "/test_seq/test_bwa2.sam"
self.reads1 = Reads()
self.reads2 = Reads()
self.reads1.load(sam1_file)
self.reads2.load(sam2_file)
self.reads1.filter_unmapped()
self.reads2.filter_unmapped()
self.genome = Genome.from_folder(self.dir + "/test_seq/dmel_genome/")
self.pairs = FragmentMappedReadPairs()
regions = self.genome.get_regions('MboI')
self.pairs.load(self.reads1, self.reads2, regions=regions)
regions.close()
def teardown_method(self, method):
self.reads1.close()
self.reads2.close()
self.genome.close()
self.pairs.close()
def test_loaded_bwamem_pairs(self):
assert self.pairs._single_count == 896
assert self.pairs._pair_count == 515
def test_pcr_duplicate_filter(self):
mask = self.pairs.add_mask_description(
'pcr_duplicate', 'Mask suspected PCR duplicates.')
pcr_duplicate_filter = PCRDuplicateFilter(pairs=self.pairs,
threshold=3)
assert len(self.pairs) == 515
self.pairs.filter(pcr_duplicate_filter)
assert len(self.pairs) == 512
class TestFragmentMappedReadPairs:
@classmethod
def setup_method(self, method):
self.dir = os.path.dirname(os.path.realpath(__file__))
sam1_file = self.dir + "/test_seq/lambda_reads1.sam"
sam2_file = self.dir + "/test_seq/lambda_reads2.sam"
self.reads1 = Reads(sam1_file)
self.reads2 = Reads(sam2_file)
self.reads1.filter_unmapped()
self.reads2.filter_unmapped()
self.genome = Genome.from_folder(self.dir + "/test_seq/lambda_genome/")
self.pairs_class = FragmentMappedReadPairs
self.pairs = self.pairs_class()
regions = self.genome.get_regions(1000)
self.pairs.load(self.reads1, self.reads2, regions=regions)
regions.close()
def teardown_method(self, method):
self.reads1.close()
self.reads2.close()
self.genome.close()
self.pairs.close()
def test_select(self):
pair = self.pairs[0]
assert isinstance(pair, FragmentReadPair)
assert pair.left.position == 40884
assert pair.right.position == 41039
assert pair.left.strand == 1
assert pair.right.strand == -1
assert isinstance(pair.left.fragment, GenomicRegion)
assert isinstance(pair.right.fragment, GenomicRegion)
assert pair.left.fragment.start == 40001
assert pair.left.fragment.end == 41000
assert pair.left.fragment.chromosome == 'gi|9626243|ref|NC_001416.1|'
assert pair.right.fragment.start == 41001
assert pair.right.fragment.end == 42000
assert pair.right.fragment.chromosome == 'gi|9626243|ref|NC_001416.1|'
pair = self.pairs[1]
assert isinstance(pair, FragmentReadPair)
assert pair.left.position == 19617
assert pair.right.position == 19736
assert pair.left.strand == 1
assert pair.right.strand == -1
assert isinstance(pair.left.fragment, GenomicRegion)
assert isinstance(pair.right.fragment, GenomicRegion)
assert pair.left.fragment.start == 19001
assert pair.left.fragment.end == 20000
assert pair.left.fragment.chromosome == 'gi|9626243|ref|NC_001416.1|'
assert pair.right.fragment.start == 19001
assert pair.right.fragment.end == 20000
assert pair.right.fragment.chromosome == 'gi|9626243|ref|NC_001416.1|'
pair = self.pairs[-1]
assert isinstance(pair, FragmentReadPair)
assert pair.left.position == 25765
assert pair.right.position == 25622
assert pair.left.strand == -1
assert pair.right.strand == 1
assert isinstance(pair.left.fragment, GenomicRegion)
assert isinstance(pair.right.fragment, GenomicRegion)
assert pair.left.fragment.start == 25001
assert pair.left.fragment.end == 26000
assert pair.left.fragment.chromosome == 'gi|9626243|ref|NC_001416.1|'
assert pair.right.fragment.start == 25001
assert pair.right.fragment.end == 26000
assert pair.right.fragment.chromosome == 'gi|9626243|ref|NC_001416.1|'
def test_iter(self):
for pair in self.pairs:
assert isinstance(pair, FragmentReadPair)
assert isinstance(pair.left, FragmentRead)
assert isinstance(pair.right, FragmentRead)
assert isinstance(pair.left.fragment, GenomicRegion)
assert isinstance(pair.right.fragment, GenomicRegion)
assert pair.left.position > 0 or pair.left.position == -1
assert pair.right.position > 0 or pair.right.position == -1
assert pair.left.strand == -1 or pair.left.strand == 1
assert pair.right.strand == -1 or pair.right.strand == 1
assert 0 < pair.left.fragment.start <= pair.left.fragment.end
assert 0 < pair.right.fragment.start <= pair.right.fragment.end
if pair.left.position > 0:
assert pair.left.fragment.start <= pair.left.position <= pair.left.fragment.end
if pair.right.position > 0:
assert pair.right.fragment.start <= pair.right.position <= pair.right.fragment.end
def test_len(self):
assert len(self.pairs) == 44
def test_single(self):
assert len(self.pairs._single) == 6
def test_auto_mindist(self):
ad = self.pairs_class._auto_dist
np.random.seed(101)
x = np.linspace(1, 100, 10)
i = [
4.0, 3.0, 1.0, 1.2, 0.4, 0.8, 0.5, 0.4, 0.6, 0.5, 0.6, 0.5, 0.5,
0.5
]
o = [
0.2, 0.4, 0.3, 0.6, 0.5, 0.4, 0.5, 0.5, 0.6, 0.5, 0.5, 0.4, 0.5,
0.5
]
b = np.random.normal(500, 200, 14).astype(int)
assert ad(x, i, b, 0.05) == 67
assert ad(x, o, b, 0.05) == 45
def test_filter_inward(self):
mask = self.pairs.add_mask_description(
'inwards',
'Mask read pairs that inward facing and closer than 100bp')
in_filter = InwardPairsFilter(minimum_distance=100, mask=mask)
assert len(self.pairs) == 44
self.pairs.filter(in_filter)
assert len(self.pairs) == 18
def test_filter_outward(self):
mask = self.pairs.add_mask_description(
'outwards',
'Mask read pairs that outward facing and closer than 100bp')
out_filter = OutwardPairsFilter(minimum_distance=100, mask=mask)
assert len(self.pairs) == 44
self.pairs.filter(out_filter)
assert len(self.pairs) == 28
def test_filter_redist(self):
mask = self.pairs.add_mask_description(
're-dist',
'Mask read pairs where one half maps more than 100bp away from both RE sites'
)
re_filter = ReDistanceFilter(maximum_distance=300, mask=mask)
assert len(self.pairs) == 44
self.pairs.filter(re_filter)
assert len(self.pairs) == 13
def test_filter_self_ligated(self):
mask = self.pairs.add_mask_description(
'self_ligated',
'Mask read pairs that represent self-ligated fragments')
self_ligation_filter = SelfLigationFilter(mask=mask)
assert len(self.pairs) == 44
self.pairs.filter(self_ligation_filter)
assert len(self.pairs) == 7
def test_get_ligation_structure_biases(self):
reads1 = Reads(self.dir +
"/../data/test_genomic/yeast.sample.chrI.1.sam")
reads2 = Reads(self.dir +
"/../data/test_genomic/yeast.sample.chrI.2.sam")
chrI = Chromosome.from_fasta(self.dir +
"/../data/test_genomic/chrI.fa")
genome = Genome(chromosomes=[chrI])
pairs = self.pairs_class()
regions = genome.get_regions('HindIII')
pairs.load(reads1, reads2, regions)
reads1.close()
reads2.close()
genome.close()
regions.close()
x, i, o, b = pairs.get_ligation_structure_biases(
sampling=200, skip_self_ligations=False)
assert len(x) == len(i) == len(o) == len(b) == 3
assert x.tolist() == [494, 4487, 19399]
assert i.tolist() == [
2.616915422885572, 0.8059701492537313, 0.6417910447761194
]
assert o.tolist() == [
0.2537313432835821, 0.24378109452736318, 0.46766169154228854
]
assert b.tolist() == [778, 412, 424]
pairs.close()
def test_re_dist(self):
read1 = FragmentRead(GenomicRegion(chromosome='chr1',
start=1,
end=1000),
position=200,
strand=-1)
assert read1.re_distance() == 199
read2 = FragmentRead(GenomicRegion(chromosome='chr1',
start=1,
end=1000),
position=990,
strand=-1)
assert read2.re_distance() == 10
def test_iterate_exclude_filters(self):
mask = self.pairs.add_mask_description(
'inwards',
'Mask read pairs that inward facing and closer than 100bp')
in_filter = InwardPairsFilter(minimum_distance=100, mask=mask)
self.pairs.filter(in_filter)
mask = self.pairs.add_mask_description(
'outwards',
'Mask read pairs that outward facing and closer than 100bp')
out_filter = OutwardPairsFilter(minimum_distance=100, mask=mask)
self.pairs.filter(out_filter)
mask = self.pairs.add_mask_description(
're-dist',
'Mask read pairs where one half maps more than 100bp away from both RE sites'
)
re_filter = ReDistanceFilter(maximum_distance=300, mask=mask)
self.pairs.filter(re_filter)
assert len(self.pairs) == 0
assert len(
list(
self.pairs.pairs(
excluded_filters=['inwards', 'outwards', 're-dist'
]))) == 44
assert len(
list(self.pairs.pairs(
excluded_filters=['inwards', 'outwards']))) == 13
assert len(
list(self.pairs.pairs(
excluded_filters=['inwards', 're-dist']))) == 28
assert len(
list(self.pairs.pairs(
excluded_filters=['outwards', 're-dist']))) == 18
assert len(list(self.pairs.pairs(excluded_filters=['inwards']))) == 11
assert len(list(self.pairs.pairs(excluded_filters=['outwards']))) == 2
assert len(list(self.pairs.pairs(excluded_filters=['re-dist']))) == 2
assert len(
list(self.pairs.pairs(
excluded_filters=[in_filter, re_filter]))) == 28
assert len(list(
self.pairs.pairs(excluded_filters=[in_filter, mask]))) == 28
assert len(list(
self.pairs.pairs(excluded_filters=[in_filter, 3]))) == 28
@pytest.mark.parametrize("gz", [False, True])
@pytest.mark.parametrize("include_filtered", [False, True])
def test_export_homer(self, gz, include_filtered, tmpdir_factory):
fn = str(
tmpdir_factory.mktemp("output").join(".tsv.gz" if gz else ".tsv"))
mask = self.pairs.add_mask_description(
'self_ligated',
'Mask read pairs that represent self-ligated fragments')
self_ligation_filter = SelfLigationFilter(mask=mask)
self.pairs.filter(self_ligation_filter)
self.pairs.to_homer(fn, include_filtered=include_filtered)
if gz:
import gzip, sys
read_handle = gzip.open(
fn, mode="rt" if sys.version_info.major == 3 else "r")
else:
read_handle = open(fn, mode="r")
with read_handle:
lines = read_handle.readlines()
assert len(lines) == 44 if include_filtered else 7