def test_getRelPos(self):
a = Interval(5, 15)
self.assertEqual(a.get_rel_pos(5), 0)
self.assertEqual(a.get_rel_pos(10), 5)
self.assertEqual(a.get_rel_pos(14), 9)
self.assertRaises(IndexError, a.get_rel_pos, 15)
def test_getAbsPos(self):
a = Interval(5, 15)
self.assertEqual(a.get_abs_pos(0), 5)
self.assertEqual(a.get_abs_pos(5), 10)
self.assertEqual(a.get_abs_pos(9), 14)
self.assertRaises(IndexError, a.get_abs_pos, 10)
def test_get_rel_pos(self):
ivl = Interval(100, 200)
self.assertEqual(0, ivl.get_rel_pos(100))
self.assertEqual(99, ivl.get_rel_pos(199))
self.assertRaises(IndexError, ivl.get_rel_pos, 99)
self.assertRaises(IndexError, ivl.get_rel_pos, 200)
def clip_read(read):
read_interval = Interval(read.pos, read.pos + len(read.seq))
try:
matches = intervals.fetch(read.rname, read_interval.start, read_interval.stop)
matches.sort(key=lambda x: len(read_interval.intersect(x)))
match = matches[-1]
except KeyError as e:
return read, 'warning, reference sequence "{}" not found'.format(str(e))
except IndexError:
return read, 'warning, no overlapping intervals'
match = Interval(match.start + offset_5p + 1, match.stop + offset_3p)
cigar = expand_cigar(read.cigar)
new_cigar = []
cigar_pos = 0
while cigar_pos < len(cigar) and cigar[cigar_pos] != 'M':
if cigar[cigar_pos] in 'MIS=X':
new_cigar.append('S')
cigar_pos += 1
ref_pos = read.pos
while cigar_pos < len(cigar) and ref_pos < match.start:
ref_pos += cigar[cigar_pos] in 'MDNS=X'
if cigar[cigar_pos] in 'MIS=X':
new_cigar.append('S')
elif cigar[cigar_pos] in 'D':
pass
else:
new_cigar.append(cigar[cigar_pos])
cigar_pos += 1
while cigar_pos < len(cigar) and cigar[cigar_pos] != 'M':
ref_pos += cigar[cigar_pos] in 'MDNS=X'
new_cigar.append(cigar[cigar_pos])
cigar_pos += 1
pos = ref_pos
while cigar_pos < len(cigar) and ref_pos < match.stop:
ref_pos += cigar[cigar_pos] in 'MDNS=X'
new_cigar.append(cigar[cigar_pos])
cigar_pos += 1
while cigar_pos < len(cigar):
if cigar[cigar_pos] in 'MIS=X':
new_cigar.append('S')
cigar_pos += 1
parts = list(read)
parts[3] = pos
parts[5] = contract_cigar(new_cigar)
return SamLine(*parts), 'clipped'
def difference(self, other, *, ignore_strand=False):
if not ignore_strand:
self._assert_same_chromosome_and_strand(other)
if not self.overlaps(other):
return Interval.INTERVAL_PAIR(self, None)
left = right = None
if self.start < other.start:
left = GenomicInterval(self.start.position, other.start.position, chromosome=self.chromosome,
strand=self.strand, data=self.data)
if other.stop < self.stop:
right = GenomicInterval(other.stop.position, self.stop.position, chromosome=self.chromosome,
strand=self.strand, data=self.data)
return Interval.INTERVAL_PAIR(left, right)
def __init__(self, iterator):
self.key_index = MultiDimensionMap([str])
self.ivl_index = MultiDimensionMap([str, Interval])
self.data = list(iterator)
for i, entry in enumerate(self.data):
self.key_index[entry.name] = i
self.ivl_index[(entry.chr, Interval(entry.start, entry.stop))] = i
def __init__(self, iterator):
self.key_index = {}
self.ivl_index = MultiDimensionMap([str, Interval])
self.data = list(iterator)
for i, entry in enumerate(self.data):
self.key_index[entry.data['name']] = i
self.ivl_index[(entry.chromosome,
Interval(entry.start.position,
entry.stop.position))] = i
def fetch(self, chr, start, stop):
return [
self.data[v] for v in self.ivl_index[chr,
Interval(start, stop)]
]
def __init__(self, iterator):
self.data = list(iterator)
self.ivl_index = MultiDimensionMap([str, Interval])
for i, variant in enumerate(self.data):
ivl = Interval(variant.position, variant.position + len(variant.data['ref']))
self.ivl_index[(variant.chromosome, ivl)] = i
def fetch(self, chr, start, stop=None):
if stop is None:
stop = start + 1
idxs = self.ivl_index[(chr, Interval(start, stop))]
return [self.data[v] for v in idxs]
def get_item(starts, stops):
return [start if start == stop else Interval(start, stop) for start, stop in zip(starts, stops)]
def __init__(self, iterator):
self.data = list(iterator)
self.ivl_index = MultiDimensionMap([str, Interval])
for i, entry in enumerate(self.data):
ivl = Interval(entry.start, entry.stop)
self.ivl_index[(entry.chr, ivl)] = i
def test_overlaps(self):
a = Interval(0, 10)
b = Interval(6, 16)
c = Interval(12, 22)
d = Interval(2, 8)
e = Interval(10, 20)
self.assertTrue(a.overlaps(b))
self.assertTrue(b.overlaps(a))
self.assertFalse(a.overlaps(c))
self.assertFalse(c.overlaps(a))
self.assertTrue(a.overlaps(d))
self.assertTrue(d.overlaps(a))
self.assertFalse(a.overlaps(e))
self.assertFalse(e.overlaps(a))
def test_touches(self):
a = Interval(0, 10)
b = Interval(6, 16)
c = Interval(12, 22)
d = Interval(2, 8)
e = Interval(10, 20)
self.assertFalse(a.touches(b))
self.assertFalse(b.touches(a))
self.assertFalse(a.touches(c))
self.assertFalse(c.touches(a))
self.assertFalse(a.touches(d))
self.assertFalse(d.touches(a))
self.assertTrue(a.touches(e))
self.assertTrue(e.touches(a))
def test_contains(self):
a = Interval(0, 10)
b = Interval(6, 16)
c = Interval(12, 22)
d = Interval(2, 8)
e = Interval(10, 20)
self.assertFalse(a.contains(b))
self.assertFalse(b.contains(a))
self.assertFalse(a.contains(c))
self.assertFalse(c.contains(a))
self.assertTrue(a.contains(d))
self.assertFalse(d.contains(a))
self.assertFalse(a.contains(e))
self.assertFalse(d.contains(e))