def test_find_subsequence_indices(self):
with captured_output() as (_, _):
# no substrings
indices = find_substring_indices("AAAAA",
'G',
offset=0,
overlap=True)
self.assertEqual([x for x in indices], [])
# yes substrings
indices = find_substring_indices("AAAAA",
'A',
offset=0,
overlap=True)
self.assertEqual([x for x in indices], [0, 1, 2, 3, 4])
indices = find_substring_indices("AAAAA",
'AA',
offset=0,
overlap=True)
self.assertEqual([x for x in indices], [0, 1, 2, 3])
# test overlap
indices = find_substring_indices("AAAAA",
'AA',
offset=0,
overlap=False)
self.assertEqual([x for x in indices], [0, 2])
# test offset
indices = find_substring_indices("ATGCATGC",
'ATGCATGC',
offset=1,
overlap=True)
self.assertEqual([x for x in indices], [1])
indices = find_substring_indices("ATGCATGC",
'ATGCATGC',
offset=1,
overlap=True)
self.assertEqual([x for x in indices], [1])
# compare with gatc modtifs
indices2 = find_substring_indices("gatcgatc",
'gatc',
offset=1,
overlap=True)
self.assertEqual([x for x in indices2], [1, 5])
# if empty string passed
indices2 = find_substring_indices("",
'gatc',
offset=1,
overlap=True)
self.assertRaises(AssertionError, indices2.__next__)
indices2 = find_substring_indices("gatcgatc",
'',
offset=1,
overlap=True)
self.assertRaises(AssertionError, indices2.__next__)
def find_gatc_motifs(sequence):
"""Generate index of 'A' within the 'GATC' motifs in a nucleotide sequence
:param sequence: since GATC motif is in DNA, expecting a DNA nucleotide sequence
:return: generator yielding index of 'A' within the 'GATC'
"""
return find_substring_indices(sequence.upper(), "GATC", offset=1)
def find_motifs_sequence_positions(sequence, motifs, overlap=False):
"""Find locations of edited nucleotide nucleotide sequence using find and replace motifs
note: we convert sequence to uppercase
:param sequence: nucleotide sequence
:param motifs: list of motif's which need to be replaced: eg [[find, replace]], [["CCAGG", "CEAGG"]]
:param overlap: boolean option to look for motif overlaps
"""
already_repaced_indexes = set()
# gather motifs
for motif_pair in motifs:
assert len(motif_pair) == 2 and type(
motif_pair
) is list, "Motifs must be structured as list of lists, even for one motif find and replace"
# find edit character and offset
offset, old_char, substitution_char = find_modification_index_and_character(
motif_pair[0], motif_pair[1])
for index in find_substring_indices(sequence.upper(),
motif_pair[0].upper(),
offset=offset,
overlap=overlap):
# make sure that there is no overlapping assignments of characters
assert index not in already_repaced_indexes, "Motifs has two different edits to a single nucleotide " \
"location. Check motifs {}".format(motifs)
already_repaced_indexes.add(index)
yield index, old_char, substitution_char
def write_file(test_seq, kmer, outfile):
with open(outfile, "w") as fh:
indices = [x+1 for x in find_substring_indices(test_seq, kmer, overlap=False)]
fh.write(kmer+"\n")
for x in indices:
fh.write(str(x)+" ")
fh.write(str(len(kmer)))
fh.write("M"+"\n")
def test_count_kmers(self):
self.assertRaises(AssertionError, count_kmers, '', 1)
self.assertRaises(AssertionError, count_kmers, 'asdf', 0)
for x in range(10):
rand_len = np.random.randint(1, 1000)
random_string = get_random_string(rand_len)
rand_k = np.random.randint(1, 10)
kmer_counts = count_kmers(random_string, rand_k)
for kmer in kmer_counts.keys():
self.assertEqual(kmer_counts[kmer], len([x for x in find_substring_indices(random_string, kmer)]),
"random_string: {} \n kmer: {}".format(random_string, kmer))