def find_overlap(s, t, min_overlap=1):
"""
Detect if s and t overlap.
Returns:
None if no overlap was detected.
0 if s is a prefix of t or t is a prefix of s.
Positive int gives index where t starts within s.
Negative int gives -index where s starts within t.
>>> find_overlap('ABCDE', 'CDE')
2
>>> find_overlap('CDE', 'ABCDEFG')
-2
>>> find_overlap('ABC', 'X') is None
True
"""
aligner = Aligner(s, max_error_rate=0)
aligner.min_overlap = min_overlap
result = aligner.locate(t)
if result is None:
return None
s_start, _, t_start, _, _, _ = result
return s_start - t_start
def test_n_wildcard_in_ref_matches_n_wildcard_in_query_back():
aligner = Aligner("NNACGT",
max_error_rate=0,
wildcard_ref=True,
flags=Where.BACK.value)
match = aligner.locate("AAANTACGTAAA")
assert match == (0, 6, 3, 9, 6, 0)
def _aligner(self, mismatch):
if not hasattr(self, "_aligner_a"):
max_err_r = (float(mismatch[0]) / len(self.index_a),
float(mismatch[1]) / len(self.index_b))
self._aligner_a = Aligner(self.index_a, max_err_r[0])
self._aligner_b = Aligner(self.index_b, max_err_r[1])
self._aligner_a.min_overlap = len(self.index_a) - mismatch[0]
self._aligner_b.min_overlap = len(self.index_b) - mismatch[1]
return self._aligner_a, self._aligner_b
def locate(reference,
query,
max_error_rate,
flags=SEMIGLOBAL,
wildcard_ref=False,
wildcard_query=False,
min_overlap=1):
aligner = Aligner(reference,
max_error_rate,
flags,
wildcard_ref,
wildcard_query,
min_overlap=min_overlap)
return aligner.locate(query)
def test_n_wildcards_not_counted_aligner_back():
ref = 'AGGNNNNNNNNNNNNNNTTC'
assert len(ref) == 20
aligner = Aligner(ref, max_error_rate=0.1, wildcard_ref=True, flags=Where.BACK.value, min_overlap=3)
assert aligner.effective_length == 6
assert aligner.locate('TTC') is None
# adapter start, adapter stop, read start, read stop
assert aligner.locate('AGG')[:4] == (0, 3, 0, 3)
assert aligner.locate('AGGCCCCCCC')[:4] == (0, 10, 0, 10)
assert aligner.locate('ATGCCCCCCC') is None
assert aligner.locate('AGGCCCCCCCCCCCCCCATC') is None
assert aligner.locate('CCC' + ref.replace('N', 'G') + 'AAA') == (0, 20, 3, 23, 20, 0)
def test_n_wildcards_not_counted_aligner_back():
ref = 'AGGNNNNNNNNNNNNNNTTC'
assert len(ref) == 20
aligner = Aligner(ref,
max_error_rate=0.1,
wildcard_ref=True,
flags=Where.BACK.value,
min_overlap=3)
assert aligner.effective_length == 6
assert aligner.locate('TTC') is None
# adapter start, adapter stop, read start, read stop
assert aligner.locate('AGG')[:4] == (0, 3, 0, 3)
assert aligner.locate('AGGCCCCCCC')[:4] == (0, 10, 0, 10)
assert aligner.locate('ATGCCCCCCC') is None
assert aligner.locate('AGGCCCCCCCCCCCCCCATC') is None
assert aligner.locate('CCC' + ref.replace('N', 'G') + 'AAA') == (0, 20, 3,
23, 20, 0)
def test_edit_environment(k, s, environment_func):
result = list(environment_func(s, k))
strings, distances, matches = zip(*result)
naive = set(naive_edit_environment(s, k))
assert len(set(strings)) == len(strings)
assert set(strings) == naive
error_rate = k / len(s) if s else 0.0
aligner = Aligner(s,
max_error_rate=error_rate,
flags=0,
min_overlap=len(s))
for t, dist, m in result:
result = aligner.locate(t)
start1, stop1, start2, stop2, matches, errors = result
assert errors == dist
assert m == matches
assert start1 == 0
assert stop1 == len(s)
assert start2 == 0
assert stop2 == len(t)
assert edit_distance(s, t) == dist
assert m <= len(s), (s, t, dist)
assert m <= len(t), (s, t, dist)
def test(self):
reference = 'CTCCAGCTTAGACATATC'
aligner = Aligner(reference, 0.1, flags=BACK)
aligner.locate('CC')
def test_100_percent_error_rate(self):
reference = 'GCTTAGACATATC'
aligner = Aligner(reference, 1.0, flags=BACK)
aligner.locate('CAA')
def test_not_only_n_wildcards(self):
reference = 'NNNNN'
with pytest.raises(ValueError) as info:
Aligner(reference, 0.1, wildcard_ref=True)
assert "only N wildcards" in info.value.args[0]
def test_100_percent_error_rate(self):
reference = 'GCTTAGACATATC'
aligner = Aligner(reference, 1.0, flags=Where.BACK.value)
aligner.locate('CAA')
def test(self):
reference = 'CTCCAGCTTAGACATATC'
aligner = Aligner(reference, 0.1, flags=Where.BACK.value)
aligner.locate('CC')
def locate(reference, query, max_error_rate, flags=SEMIGLOBAL, wildcard_ref=False,
wildcard_query=False, min_overlap=1):
aligner = Aligner(reference, max_error_rate, flags, wildcard_ref, wildcard_query, min_overlap=min_overlap)
return aligner.locate(query)
def test_find_empty_in_empty(self):
aligner = Aligner("", 0, flags=0, min_overlap=0)
result = aligner.locate("")
assert (0, 0, 0, 0, 0, 0) == result
