def test_partitions_bad_interval_size_raises(self):
# list() is necessary to force the generator to execute.
with self.assertRaisesRegexp(ValueError, 'max_size'):
list(
ranges.RangeSet([ranges.make_range('chrM', 0,
100)]).partition(-10))
with self.assertRaisesRegexp(ValueError, 'max_size'):
list(
ranges.RangeSet([ranges.make_range('chrM', 0,
100)]).partition(0))
def test_envelops(self):
start_ix = 5
end_ix = 10
start_ix2 = end_ix + 1
end_ix2 = end_ix + 5
range_set = ranges.RangeSet([
ranges.make_range('chr1', start_ix, end_ix),
ranges.make_range('chr1', start_ix2, end_ix2)
])
# No start position before the first start range is enveloped.
for i in range(start_ix):
self.assertFalse(range_set.envelops('chr1', i, start_ix + 1))
# All regions within a single record are enveloped.
for six in range(start_ix, end_ix):
for eix in range(six, end_ix + 1):
self.assertTrue(range_set.envelops('chr1', six, eix),
'chr1 {} {} not enveloped'.format(six, eix))
# Bridging across two ranges is not enveloped.
for six in range(start_ix, end_ix):
for eix in range(start_ix2, end_ix2 + 1):
self.assertFalse(range_set.envelops('chr1', six, eix))
# Other chromosome is not spanned.
self.assertFalse(range_set.envelops('chr2', start_ix, start_ix + 1))
def test_overlaps_variant_with_ranges(self):
variant = variants_pb2.Variant(reference_name='chr2', start=10, end=11)
range_set = ranges.RangeSet([ranges.make_range('chr1', 0, 5)])
with mock.patch.object(range_set, 'overlaps') as mock_overlaps:
mock_overlaps.return_value = True
self.assertEqual(range_set.variant_overlaps(variant), True)
mock_overlaps.assert_called_once_with('chr2', 10)
def test_detector_ranges(self):
test_ranges = [
ranges.make_range('chr1', 0, 5),
ranges.make_range('chr1', 8, 10),
ranges.make_range('chr1', 12, 13),
ranges.make_range('chr2', 2, 5),
]
range_set = ranges.RangeSet(test_ranges)
self.assertEqual(bool(range_set), True)
self.assertEqual(len(range_set), 4)
self.assertEqual(range_set.overlaps('chr1', 0), True)
self.assertEqual(range_set.overlaps('chr1', 1), True)
self.assertEqual(range_set.overlaps('chr1', 2), True)
self.assertEqual(range_set.overlaps('chr1', 3), True)
self.assertEqual(range_set.overlaps('chr1', 4), True)
self.assertEqual(range_set.overlaps('chr1', 5), False)
self.assertEqual(range_set.overlaps('chr1', 6), False)
self.assertEqual(range_set.overlaps('chr1', 7), False)
self.assertEqual(range_set.overlaps('chr1', 8), True)
self.assertEqual(range_set.overlaps('chr1', 9), True)
self.assertEqual(range_set.overlaps('chr1', 10), False)
self.assertEqual(range_set.overlaps('chr1', 11), False)
self.assertEqual(range_set.overlaps('chr1', 12), True)
self.assertEqual(range_set.overlaps('chr1', 13), False)
self.assertEqual(range_set.overlaps('chr1', 100), False)
self.assertEqual(range_set.overlaps('chr1', 1000), False)
self.assertEqual(range_set.overlaps('chr2', 0), False)
self.assertEqual(range_set.overlaps('chr2', 1), False)
self.assertEqual(range_set.overlaps('chr2', 2), True)
self.assertEqual(range_set.overlaps('chr2', 3), True)
self.assertEqual(range_set.overlaps('chr2', 4), True)
self.assertEqual(range_set.overlaps('chr2', 5), False)
self.assertEqual(range_set.overlaps('chr2', 6), False)
self.assertEqual(range_set.overlaps('chr3', 3), False)
def test_partition_of_multiple_intervals(self, interval_size, expected):
rangeset = ranges.RangeSet([
ranges.make_range('1', 0, 10),
ranges.make_range('1', 20, 40),
ranges.make_range('1', 45, 50),
])
self.assertCountEqual([ranges.make_range(*args) for args in expected],
rangeset.partition(interval_size))
def test_partitions(self, interval_size, expected):
rangeset = ranges.RangeSet([
ranges.make_range('chrM', 0, 100),
ranges.make_range('chr1', 0, 76),
ranges.make_range('chr2', 0, 121),
])
self.assertEqual([ranges.make_range(*args) for args in expected],
list(rangeset.partition(interval_size)))
def test_intersection(self, regions, expected):
regions_list = [ranges.RangeSet.from_regions(r) for r in regions]
copies = [ranges.RangeSet(rs) for rs in regions_list]
# Check that the intersection is as expected.
self.assertCountEqual(ranges.RangeSet.from_regions(expected),
regions_list[0].intersection(*regions_list[1:]))
# Check that no one was modified.
for pre, post in zip(copies, regions_list):
self.assertCountEqual(pre, post)
def test_rangeset_iteration_order(self):
contigs = [
reference_pb2.ContigInfo(name='c', n_bases=100, pos_in_fasta=0),
reference_pb2.ContigInfo(name='b', n_bases=121, pos_in_fasta=2),
reference_pb2.ContigInfo(name='a', n_bases=76, pos_in_fasta=1),
]
unsorted = ranges.parse_literals(
['a:10', 'c:20', 'b:30', 'b:10-15', 'a:5'])
# Iteration order over a RangeSet instantiated with a contigs list is
# determined by pos_in_fasta, start, end.
range_set_with_contigs = ranges.RangeSet(unsorted, contigs)
self.assertEqual(
ranges.parse_literals(['c:20', 'a:5', 'a:10', 'b:10-15', 'b:30']),
[range_ for range_ in range_set_with_contigs])
# For a RangeSet instantiated *without* a contig map, the iteration order
# is determined by reference_name, start, end.
range_set_no_contigs = ranges.RangeSet(unsorted)
self.assertEqual(
ranges.parse_literals(['a:5', 'a:10', 'b:10-15', 'b:30', 'c:20']),
[range_ for range_ in range_set_no_contigs])
def test_unrecognized_contig_triggers_exception(self):
with self.assertRaises(ValueError):
_ = ranges.RangeSet([ranges.make_range('bogus_chromosome', 1, 10)],
_TEST_CONTIGS)
def test_detector_no_ranges(self):
range_set = ranges.RangeSet()
# don't have any ranges by default
self.assertEqual(bool(range_set), False)
# make sure we can call overlaps without any ranges
self.assertFalse(range_set.overlaps('chr1', 10))
def test_overlaps_variant_empty_range(self):
variant = variants_pb2.Variant(reference_name='chr2', start=10, end=11)
empty_set = ranges.RangeSet()
self.assertEqual(
empty_set.variant_overlaps(variant, empty_set_return_value='foo'),
'foo')