def test_find_max_overlapping_returns_least_index(self):
query_range = ranges.make_range('1', 0, 10)
search_ranges = [
ranges.make_range('1', 0, 5),
ranges.make_range('1', 5, 10)
]
for to_search in [search_ranges, list(reversed(search_ranges))]:
self.assertEqual(0, ranges.find_max_overlapping(query_range, to_search))
def test_find_max_overlapping_allows_unordered_search_ranges(self):
query_range = ranges.make_range('1', 4, 12)
search_ranges = [
ranges.make_range('1', 0, 10),
ranges.make_range('1', 10, 20),
ranges.make_range('1', 12, 20)
]
max_overlapping_range = search_ranges[0]
for permutated_ranges in itertools.permutations(search_ranges):
self.assertEqual(
permutated_ranges.index(max_overlapping_range),
ranges.find_max_overlapping(query_range, permutated_ranges))
def assign_reads_to_assembled_regions(assembled_regions, reads):
"""Assign each read to the maximally overlapped window.
Args:
assembled_regions: list[AssemblyRegion], list of AssemblyRegion to assign
reads to. Does not assume AssemblyRegion are sorted.
reads: iterable[learning.genomics.genomics.Read], to be processed. Does
not assume the reads are sorted.
Returns:
[AssemblyRegion], information on assigned reads for each assembled region.
list[learning.genomics.genomics.Read], the list of unassigned reads.
"""
regions = [ar.region for ar in assembled_regions]
unassigned_reads = []
for read in reads:
read_range = utils.read_range(read)
window_i = ranges.find_max_overlapping(read_range, regions)
if window_i is not None:
assembled_regions[window_i].add_read(read)
else:
unassigned_reads.append(read)
return unassigned_reads
def test_find_max_overlapping(self, query_range, search_ranges, expected): actual = ranges.find_max_overlapping(query_range, search_ranges) self.assertEqual(expected, actual)