def test_lt(self):
self.assertLess(self.position2, self.position1)
self.assertGreater(self.position1, self.position2)
self.assertLess(self.position1, self.position3)
chr5_position = Position(chromosome="chr5", coordinate=5, strand=Strand.FORWARD)
chr10_position = Position(chromosome="chr10", coordinate=1, strand=Strand.REVERSE)
self.assertLess(self.position1, self.position4)
self.assertLess(chr5_position, chr10_position)
def test_creation(self):
position3 = Position(chromosome="chr1", coordinate=1, strand=Strand.FORWARD)
position4 = Position(chromosome="chr2", coordinate=2, strand=Strand.FORWARD)
position5 = Position(chromosome="chr1", coordinate=0, strand=Strand.FORWARD)
for pos in [position3, position4, position5]:
with self.assertRaises(ValueError):
Segment(start_position=self.position1, end_position=pos)
Segment(start_position=self.position1, end_position=self.position2)
def adjacencies_mergeable(adjacency1, adjacency2, max_distance=500, separate_distances=True):
distance11 = Position.non_hap_distance_strand_specific(pos1=adjacency1.position1, pos2=adjacency2.position1)
distance12 = Position.non_hap_distance_strand_specific(pos1=adjacency1.position2, pos2=adjacency2.position2)
distance21 = Position.non_hap_distance_strand_specific(pos1=adjacency1.position2, pos2=adjacency2.position1)
distance22 = Position.non_hap_distance_strand_specific(pos1=adjacency1.position1, pos2=adjacency2.position2)
if not separate_distances:
return min(distance11 + distance12, distance21 + distance22) <= max_distance
case1 = distance11 <= max_distance and distance12 <= max_distance
case2 = distance21 <= max_distance and distance22 <= max_distance
return case1 or case2
def test_eq(self):
non_position = "?"
eq_position1 = Position(chromosome="chr1", coordinate=1, strand=Strand.FORWARD)
self.assertNotEqual(self.position1, self.position2)
self.assertNotEqual(self.position1, self.position3)
self.assertNotEqual(self.position1, self.position4)
self.assertNotEqual(self.position1, non_position)
self.assertEqual(self.position1, eq_position1)
def get_merged_adjacencies(self, merged_template="{cnt}_merged"):
result = []
processed_ids = set()
cnt = 0
for lp, rp in self.merged_position_pairs:
lp_chr, lp_strand, lp_coord = lp
rp_chr, rp_strand, rp_coord = rp
lp_ids = self.get_adjs_ids_by_position(chromosome=lp_chr, strand=lp_strand, coordinate=lp_coord)
rp_ids = self.get_adjs_ids_by_position(chromosome=rp_chr, strand=rp_strand, coordinate=rp_coord)
common_ids = lp_ids & rp_ids
assert len(common_ids) > 0
assert len(common_ids & processed_ids) == 0
processed_ids.update(common_ids)
pos1 = Position(chromosome=lp_chr, coordinate=lp_coord, strand=lp_strand)
pos2 = Position(chromosome=rp_chr, coordinate=rp_coord, strand=rp_strand)
adj = Adjacency(position1=pos1, position2=pos2, extra={
EXTERNAL_NA_ID: merged_template.format(cnt=cnt),
self.origin_ids_field: ",".join(common_ids),
})
result.append(adj)
cnt += 1
return result
def setUp(self):
self.p1 = Position(chromosome="1", coordinate=1, strand=Strand.REVERSE)
self.p2 = Position(chromosome="1", coordinate=2, strand=Strand.FORWARD)
self.p3 = Position(chromosome="1", coordinate=3, strand=Strand.REVERSE)
self.p4 = Position(chromosome="1", coordinate=4, strand=Strand.FORWARD)
self.p5 = Position(chromosome="1", coordinate=5, strand=Strand.REVERSE)
self.p6 = Position(chromosome="1", coordinate=6, strand=Strand.FORWARD)
self.p7 = Position(chromosome="2", coordinate=1, strand=Strand.REVERSE)
self.p8 = Position(chromosome="2", coordinate=2, strand=Strand.FORWARD)
self.p9 = Position(chromosome="2", coordinate=3, strand=Strand.REVERSE)
self.p10 = Position(chromosome="2",
coordinate=4,
strand=Strand.FORWARD)
self.p11 = Position(chromosome="2",
coordinate=5,
strand=Strand.REVERSE)
self.p12 = Position(chromosome="2",
coordinate=6,
strand=Strand.FORWARD)
self.s1 = Segment(start_position=self.p1, end_position=self.p2)
self.s2 = Segment(start_position=self.p3, end_position=self.p4)
self.s3 = Segment(start_position=self.p5, end_position=self.p6)
self.s4 = Segment(start_position=self.p7, end_position=self.p8)
self.s5 = Segment(start_position=self.p9, end_position=self.p10)
self.s6 = Segment(start_position=self.p11, end_position=self.p12)
def setUp(self):
self.position1 = Position(chromosome="chr1", coordinate=1, strand=Strand.REVERSE)
self.position2 = Position(chromosome="chr1", coordinate=2, strand=Strand.FORWARD)
def test_empty_extra_creation(self):
self.assertDictEqual(Position(chromosome="chrom1", coordinate=1, strand=Strand.FORWARD).extra, {})
def refined_segments(segments,
additional_positions=None,
additional_positions_by_chrs=None):
fragments = deepcopy(segments)
if additional_positions is None:
additional_positions = []
if additional_positions_by_chrs is None:
additional_positions_by_chrs = defaultdict(list)
refined_segments = []
segments_ids_mapping = defaultdict(list)
source_fragments_by_chrs = defaultdict(list)
for fragment in fragments:
source_fragments_by_chrs[fragment.chromosome].append(fragment)
for position in additional_positions:
additional_positions_by_chrs[position.chromosome].append(position)
for chr_name in list(source_fragments_by_chrs.keys()):
source_fragments_by_chrs[chr_name] = sorted(
source_fragments_by_chrs[chr_name],
key=lambda s: (s.start_coordinate, s.end_coordinate))
if not sorted_segments_donot_overlap(
segments=source_fragments_by_chrs[chr_name]):
raise ValueError(
"Some segments overlap on chromosome {chr_name}.".format(
chr_name=chr_name))
for chr_name in list(additional_positions_by_chrs.keys()):
additional_positions_by_chrs[chr_name] = sorted(
additional_positions_by_chrs[chr_name],
key=lambda p: (p.coordinate, p.strand))
for chr_name in source_fragments_by_chrs:
chr_fragments = iter(source_fragments_by_chrs[chr_name])
chr_positions = iter(additional_positions_by_chrs[chr_name])
current_fragment = next(chr_fragments, None)
current_position = next(chr_positions, None)
current_segment = deepcopy(current_fragment)
refined_segments.append(current_segment)
new_to_old = []
while current_fragment is not None and current_position is not None:
if current_position.coordinate < current_fragment.start_coordinate:
current_position = next(chr_positions, None)
elif current_position.coordinate == current_segment.start_coordinate and current_position.strand == Strand.REVERSE:
current_position = next(chr_positions, None)
elif current_position.coordinate == current_segment.end_coordinate and current_position.strand == Strand.FORWARD:
current_position = next(chr_positions, None)
elif current_position.coordinate <= current_fragment.end_coordinate:
if current_position.strand == Strand.FORWARD:
left_partition_coordinate = current_position.coordinate
else:
left_partition_coordinate = current_position.coordinate - 1
right_partition_coordinate = left_partition_coordinate + 1
new_end_position = Position(
chromosome=chr_name,
coordinate=left_partition_coordinate,
strand=Strand.FORWARD)
new_start_position = Position(
chromosome=chr_name,
coordinate=right_partition_coordinate,
strand=Strand.REVERSE)
current_segment.end_position = new_end_position
new_to_old.append(current_segment.stable_id_non_hap)
current_segment = deepcopy(current_fragment)
current_segment.start_position = new_start_position
refined_segments.append(current_segment)
current_position = next(chr_positions, None)
elif current_position.coordinate > current_fragment.end_coordinate:
new_to_old.append(current_segment.stable_id_non_hap)
current_fragment_id = current_fragment.stable_id_non_hap
for sid in new_to_old:
segments_ids_mapping[current_fragment_id].append(sid)
segments_ids_mapping[sid].append(current_fragment_id)
current_fragment = next(chr_fragments, None)
current_segment = deepcopy(current_fragment)
if current_fragment is not None:
refined_segments.append(current_segment)
new_to_old = []
else:
raise ValueError("Something went wrong")
if current_fragment is not None and current_segment is not None:
new_to_old.append(current_segment.stable_id_non_hap)
current_fragment_id = current_fragment.stable_id_non_hap
for sid in new_to_old:
segments_ids_mapping[current_fragment_id].append(sid)
segments_ids_mapping[sid].append(current_fragment_id)
current_fragment = next(chr_fragments, None)
current_segment = deepcopy(current_fragment)
while current_fragment is not None:
sid = current_segment.stable_id_non_hap
segments_ids_mapping[sid].append(sid)
refined_segments.append(current_segment)
current_fragment = next(chr_fragments, None)
current_segment = deepcopy(current_fragment)
return refined_segments, segments_ids_mapping