def test_score_threshold_rule(kmer_alphabet, ref_kmer, threshold):
"""
Test if the similar k-mers given by :class:`ScoreThresholdRule`
are equal to k-mers generated by a brute-force approach.
"""
matrix = align.SubstitutionMatrix.std_protein_matrix()
ref_kmer_sequence = seq.ProteinSequence()
ref_kmer_sequence.code = kmer_alphabet.split(ref_kmer)
ref_sim_kmer_set = set()
# Iterate through all possible k-mers
for kmer in range(len(kmer_alphabet)):
kmer_sequence = seq.ProteinSequence()
kmer_sequence.code = kmer_alphabet.split(kmer)
score = align.align_ungapped(
ref_kmer_sequence, kmer_sequence, matrix, score_only=True
)
# Add k-mer to list if the threshold score is reached
if score >= threshold:
ref_sim_kmer_set.add(kmer)
test_rule = align.ScoreThresholdRule(matrix, threshold)
test_sim_kmer_set = set(test_rule.similar_kmers(kmer_alphabet, ref_kmer))
assert test_sim_kmer_set == ref_sim_kmer_set
def test_invalid_kmer(kmer_alphabet, invalid_kmer):
"""
Expect an error when an invalid k-mer code is given as input.
"""
test_rule = align.ScoreThresholdRule(
align.SubstitutionMatrix.std_protein_matrix(), 0
)
with pytest.raises(seq.AlphabetError):
test_rule.similar_kmers(kmer_alphabet, invalid_kmer)
def test_invalid_matrix(kmer_alphabet):
"""
Expect an error when the k-mer alphabet is incompatible with the
substitution matrix.
"""
test_rule = align.ScoreThresholdRule(
align.SubstitutionMatrix.std_nucleotide_matrix(), 0
)
with pytest.raises(ValueError):
test_rule.similar_kmers(kmer_alphabet, 0)
def _identity_rule(alphabet):
score_matrix = np.full((len(alphabet),) * 2, -1, dtype=int)
np.fill_diagonal(score_matrix, 0)
matrix = align.SubstitutionMatrix(alphabet, alphabet, score_matrix)
rule = align.ScoreThresholdRule(matrix, 0)
return rule