def test_global_pairwise_align_dtype_mismatch(self):
with self.assertRaisesRegex(TypeError, r"same dtype: 'DNA' != 'RNA'"):
global_pairwise_align(DNA('ACGT'), TabularMSA([RNA('ACGU')]), 1.0,
1.0, {})
with self.assertRaisesRegex(TypeError, r"same dtype: 'DNA' != 'RNA'"):
global_pairwise_align(TabularMSA([DNA('ACGT')]),
TabularMSA([RNA('ACGU')]), 1.0, 1.0, {})
def test_global_pairwise_align_dtype_mismatch(self):
with self.assertRaisesRegex(TypeError,
"same dtype: 'DNA' != 'RNA'"):
global_pairwise_align(DNA('ACGT'), TabularMSA([RNA('ACGU')]),
1.0, 1.0, {})
with self.assertRaisesRegex(TypeError,
"same dtype: 'DNA' != 'RNA'"):
global_pairwise_align(TabularMSA([DNA('ACGT')]),
TabularMSA([RNA('ACGU')]),
1.0, 1.0, {})
def proteinAlign(seq1,
seq2,
gap_open_penalty,
gap_extend_penalty,
local=False):
seq1 = seq1.upper()
seq2 = seq2.upper()
if local:
aln, score, _ = local_pairwise_align(Protein(seq1), Protein(seq2),
gap_open_penalty,
gap_extend_penalty, blosum50)
else:
aln, score, _ = global_pairwise_align(Protein(seq1),
Protein(seq2),
gap_open_penalty,
gap_extend_penalty,
blosum50,
penalize_terminal_gaps=True)
response = {
'aln1':
str(aln[0]),
'aln2':
str(aln[1]),
'score':
score,
'similarity':
float('{:.2f}'.format(aln[0].match_frequency(aln[1], relative=True) *
100))
}
return response
def pairwise_align_and_merge_sequences(self, input1, input2):
output = []
if len(input1) > 0 and len(input2) > 0:
sub_matrix = ska.make_identity_substitution_matrix(
4, -2, alphabet=self.SignSequence.definite_chars
) # match_score, mismatch_score
x = self.SignSequence(input1.replace(' ', '*'))
y = self.SignSequence(input2.replace(' ', '*'))
alignment, score, start_end_positions = ska.global_pairwise_align(
x, y, 4, 1, substitution_matrix=sub_matrix)
# print('score %d' % score)
# print(str(alignment[0]))
# print(str(alignment[1]))
output = self.merge_sequences(str(alignment[0]), str(alignment[1]))
elif len(input1) > 0 and len(input2) == 0:
output = []
for char in input1:
output.append([char])
elif len(input2) > 0 and len(input1) == 0:
output = []
for char in input2:
output.append([char])
else:
output.append([""])
return output
def build_compact_global_mask(refs):
masks = build_masks(refs)
alignment = global_pairwise_align(CustomSequence(masks[0]),
CustomSequence(masks[1]),
gap_open_penalty=1,
gap_extend_penalty=0,
substitution_matrix=subst_matrix)
print(alignment)
global_mask = '^(?i)(' + '|'.join(
masks
) + ')|(_Hearing_IS|_Complaint_IS|_Settlement_IS|_Verdict_IS|_Withdrawal_IS)\\?$'
return global_mask, alignment
def custom_align(target, cluster):
hashes = generate_hash(target, cluster)
coded_target = code_seq(target, hashes)
coded_cluster = code_seq(cluster, hashes)
print('Coded cluster: ', coded_cluster)
class CustomSequence(GrammaredSequence):
@classproperty
def degenerate_map(cls):
return {}
@classproperty
def definite_chars(cls):
return set([hashes[k] for k in hashes])
@classproperty
def default_gap_char(cls):
return '-'
@classproperty
def gap_chars(cls):
return set('-.')
target_obj = CustomSequence(coded_target)
cluster_obj = CustomSequence(coded_cluster)
substitution_matrix = make_identity_substitution_matrix(
match_score=1,
mismatch_score=-1,
alphabet=letters
)
alignment = global_pairwise_align(
target_obj,
cluster_obj,
gap_open_penalty=1,
gap_extend_penalty=1,
substitution_matrix=substitution_matrix
)
return alignment
def test_global_pairwise_align_custom_alphabet_nondegenerate_chars(self):
custom_substitution_matrix = make_identity_substitution_matrix(
1, -1, alphabet=CustomSequence.nondegenerate_chars)
custom_msa, custom_score, custom_start_end = global_pairwise_align(
CustomSequence("WXYZ"), CustomSequence("WXYYZZ"),
10.0, 5.0, custom_substitution_matrix)
# Expected values computed by running an equivalent alignment using the
# DNA alphabet with the following mapping:
#
# W X Y Z
# | | | |
# A C G T
#
self.assertEqual(custom_msa, TabularMSA([CustomSequence('WXYZ^^'),
CustomSequence('WXYYZZ')]))
self.assertEqual(custom_score, 2.0)
self.assertEqual(custom_start_end, [(0, 3), (0, 5)])
def test_global_pairwise_align_invalid_type(self):
with self.assertRaisesRegex(TypeError,
"GrammaredSequence.*"
"TabularMSA.*'Sequence'"):
global_pairwise_align(DNA('ACGT'), Sequence('ACGT'), 1.0, 1.0, {})
def test_global_pairwise_align_invalid_type(self):
with six.assertRaisesRegex(self, TypeError,
"IUPACSequence.*TabularMSA.*'Sequence'"):
global_pairwise_align(DNA('ACGT'), Sequence('ACGT'), 1.0, 1.0, {})
def msa_alignment_skbio(text_1, text_2, text_3):
from skbio.alignment import global_pairwise_align, make_identity_substitution_matrix
from multi_sequence_alignment.scikit_custom_sequence_ocr import CustomSequence
#
gap_open_penalty = 1
gap_extend_penalty = 1
# substitution_matrix_b50 =CustomSequence.blosum50 # just an example
try:
"""
alignment, score, start_end_positions = local_pairwise_align_ssw(
DNA("ACTAAGGCTCTCTACCCCTCTCAGAGA"),
DNA("ACTAAGGCTCCTAACCCCCTTTTCTCAGA")
)
"""
# todo sequence/_sequence.py is missing proper encoding, this has to fix to make this work completely atm workaround: replace non ascii with '?'
# also sequence/_grammared_sequence.py
# cs1 = CustomSequence("Hallo das ist ein Test überkrass")
# cs2 = CustomSequence("H4llo das ist Test überkraass")
cs1 = CustomSequence(text_1)
cs2 = CustomSequence(text_2)
cs3 = CustomSequence(text_3)
# substitution_matrix_unity = cs2.create_unity_sequence_matrix()
substitution_matrix_equal = make_identity_substitution_matrix(
1, -1, cs2.create_charset_string())
# alignment, score, start_end_positions = local_pairwise_align(cs1, cs2, gap_open_penalty, gap_extend_penalty, substitution_matrix_unity)
alignment12, score12, start_end_positions12 = global_pairwise_align(
cs1, cs2, gap_open_penalty, gap_extend_penalty,
substitution_matrix_equal)
alignment23, score23, start_end_positions23 = global_pairwise_align(
cs2, cs3, gap_open_penalty, gap_extend_penalty,
substitution_matrix_equal)
# alignment3, score3, start_end_positions3 = global_pairwise_align("Hallo das ist ein Test", "H4llo das ist Test", gap_open_penalty, gap_extend_penalty, substitution_matrix_equal)
# res_one_1, res_two_1 = MsaHandler.compare(list_one, list_two)
# res_two_2, res_three_2 = MsaHandler.compare(list_two, list_three)
res_one_1 = str(alignment12._seqs[0])
res_two_1 = str(alignment12._seqs[1])
res_two_2 = str(alignment23._seqs[0])
res_three_2 = str(alignment23._seqs[1])
list_res_one_1 = list(res_one_1)
list_res_two_1 = list(res_two_1)
list_res_two_2 = list(res_two_2)
list_res_three_2 = list(res_three_2)
list_pivot_msa = None
pivot_msa = None
if len(list_res_two_1) >= len(list_res_two_2):
# if len(list_res_two_1) > len(list_res_two_2):
list_pivot_msa = list_res_two_1
pivot_msa = res_two_1
else:
list_pivot_msa = list_res_two_2
pivot_msa = res_two_2
print(len(res_one_1), res_one_1)
print(len(pivot_msa), pivot_msa)
print(len(res_three_2), res_three_2)
# if res_one_1.__contains__("Sitz:") is True:
# print("asd")
res_one_1_filled = "test" #MsaHandler.fillup_wildcarded_result(res_one_1, pivot_msa, '@')
res_three_2_filled = "test" # MsaHandler.fillup_wildcarded_result(res_three_2, pivot_msa, '@')
res_final_1 = res_one_1_filled
res_final_2 = pivot_msa
res_final_3 = res_three_2
#res_final_3 = res_three_2_filled
return res_final_1, res_final_2, res_final_3
except Exception as ex:
tr = inspect.trace()
print("Exception raised in %s" % tr[-1][3])
