def test_should_remove_gap_case_c(self):
# setup
problem = MSA(score_list=[])
problem.identifiers = ['seq1']
problem.number_of_variables = 1
msa = MSASolution(problem, msa=[('seq1', 'AB----CD-E-')])
msa.remove_gap_from_sequence(0, 3)
self.assertEqual(['AB---CD-E-'],
msa.decode_alignment_as_list_of_sequences())
msa.remove_gap_from_sequence(0, 3)
self.assertEqual(['AB--CD-E-'],
msa.decode_alignment_as_list_of_sequences())
msa.remove_gap_from_sequence(0, 8)
self.assertEqual(['AB--CD-E'],
msa.decode_alignment_as_list_of_sequences())
def test_should_fill_sequences_with_gaps_to_reach_the_max_sequence_length(self):
# setup
problem = MSA(score_list=[])
problem.identifiers = ['a', 'b']
problem.number_of_variables = 2
msa_1 = MSASolution(problem, msa=[('a', '-----GE'), ('b', 'KWPFFQEAQK')])
msa_2 = MSASolution(problem, msa=[('a', '-----GE'), ('b', 'KWPFFQEAQK')])
msa_3 = MSASolution(problem, msa=[('a', '-'), ('b', 'ABC')])
crossover = SPXMSA(probability=1.0, remove_gap_columns=False)
# run
crossover.fill_sequences_with_gaps_to_reach_the_max_sequence_length(msa_1, 10, [-1, -1])
crossover.fill_sequences_with_gaps_to_reach_the_max_sequence_length(msa_2, 10, [-1, 5])
crossover.fill_sequences_with_gaps_to_reach_the_max_sequence_length(msa_3, 5, [-1, 1])
# check
self.assertEqual(["-----G---E", "KWPFFQEAQK"], msa_1.decode_alignment_as_list_of_sequences())
self.assertEqual(["-----G---E", "KWPFFQEAQK"], msa_2.decode_alignment_as_list_of_sequences())
self.assertEqual(["-----", "AB--C"], msa_3.decode_alignment_as_list_of_sequences())
def evaluate(self, solution: MSASolution) -> MSASolution:
solution.remove_full_of_gaps_columns()
sequences = solution.decode_alignment_as_list_of_sequences()
for i, score in enumerate(self.score_list):
solution.objectives[i] = score.compute(sequences)
if not score.is_minimization():
solution.objectives[i] = -solution.objectives[i]
return solution
def test_should_return_original_sequences(self):
# setup
problem = MSA(score_list=[])
problem.identifiers = ['seq1', 'seq2', 'seq3']
problem.number_of_variables = 3
msa = MSASolution(problem,
msa=[('seq1', 'AC---TGAC'), ('seq2', 'AT--CT--C'),
('seq3', 'AAC---TGC')])
# check
self.assertEqual(['AC---TGAC', 'AT--CT--C', 'AAC---TGC'],
msa.decode_alignment_as_list_of_sequences())
def test_should_remove_all_gap_columns_case_d(self):
# setup
problem = MSA(score_list=[])
problem.identifiers = ['seq1', 'seq2']
problem.number_of_variables = 2
msa = MSASolution(problem,
msa=[('seq1', 'AB--CDE-'), ('seq2', 'AB--CD-E')])
msa.remove_full_of_gaps_columns()
# check
self.assertEqual(['ABCDE-', 'ABCD-E'],
msa.decode_alignment_as_list_of_sequences())
def test_should_remove_all_gap_columns_case_b(self):
# setup
problem = MSA(score_list=[])
problem.identifiers = ['seq1', 'seq2', 'seq3']
problem.number_of_variables = 3
msa = MSASolution(problem,
msa=[('seq1', 'AC--T--GC'), ('seq2', 'AC-----AC'),
('seq3', 'A---C--AC')])
msa.remove_full_of_gaps_columns()
# check
self.assertEqual(['ACTGC', 'AC-AC', 'A-CAC'],
msa.decode_alignment_as_list_of_sequences())
def test_should_remove_gap(self):
# setup
problem = MSA(score_list=[])
problem.identifiers = ['seq1', 'seq2', 'seq3']
problem.number_of_variables = 3
msa = MSASolution(problem,
msa=[('seq1', 'AC---TGAC'), ('seq2', 'AC---TGAC'),
('seq3', 'AC---TGAC')])
msa.remove_gap_from_sequence(0, 2)
msa.remove_gap_from_sequence(1, 2)
msa.remove_gap_from_sequence(2, 2)
# check
self.assertEqual(['AC--TGAC', 'AC--TGAC', 'AC--TGAC'],
msa.decode_alignment_as_list_of_sequences())
def test_should_merge_gaps_groups_case_b(self):
# setup
problem = MSA(score_list=[])
problem.identifiers = ['seq1']
problem.number_of_variables = 1
aln_seq = [('seq1', 'ACTGAC')]
msa = MSASolution(problem, msa=aln_seq)
msa.gaps_groups[0] = [2, 4, 4, 8, 8, 10]
self.assertEqual(["AC-----------TGAC"],
msa.decode_alignment_as_list_of_sequences())
# run
msa.merge_gaps_groups()
# check
self.assertEqual([2, 10], msa.gaps_groups[0])
def test_should_merge_gaps_groups(self):
# setup
problem = MSA(score_list=[])
problem.identifiers = ['seq1', 'seq2']
problem.number_of_variables = 2
aln_seq = [('seq1', 'ACTGAC'), ('seq2', 'ATCTC')]
msa = MSASolution(problem, msa=aln_seq)
msa.gaps_groups[0] = [2, 4, 4, 5]
msa.gaps_groups[1] = [2, 4, 5, 8]
self.assertEqual(["AC-----TGAC", "AT-------CTC"],
msa.decode_alignment_as_list_of_sequences())
# run
msa.merge_gaps_groups()
# check
self.assertEqual([2, 5], msa.gaps_groups[0])
self.assertEqual([2, 8], msa.gaps_groups[1])
