def test_protein_subsequence_does_not_overlap_substitution():
# testing that we got the correct properties for case where "SIINFEKL" was
# mutated into "SIINFEQL" and then sliced to keep just "FEQL"
p = ProteinSequence(amino_acids="SIINFEQL",
contains_mutation=True,
mutation_start_idx=len("SIINFE"),
mutation_end_idx=len("SIINFEQ"),
ends_with_stop_codon=True,
frameshift=False,
translations=[])
p2 = p.subsequence(0, len("SIIN"))
eq_(p2.amino_acids, "SIIN")
eq_(p2.contains_mutation, False)
eq_(p2.contains_deletion, False)
eq_(p2.frameshift, False)
eq_(p2.ends_with_stop_codon, False)
eq_(p2.num_mutant_amino_acids, 0)
eq_(p2.mutant_amino_acids, "")
eq_(len(p2), 4)
def test_protein_subsequence_does_not_overlap_deletion():
# testing that we got correct properties for the case
# where "SIINFEKL" was mutated into "SIINFEL" (by a deletion of "K")
# and then we took the subsequence "SIINFE"
p = ProteinSequence(amino_acids="SIINFEL",
contains_mutation=True,
mutation_start_idx=len("SIINFE"),
mutation_end_idx=len("SIINFE"),
ends_with_stop_codon=True,
frameshift=False,
translations=[])
p2 = p.subsequence(None, len("SIINFE"))
eq_(len(p2), 6)
eq_(p2.amino_acids, "SIINFE")
eq_(p2.contains_deletion, False)
eq_(p2.contains_mutation, False)
eq_(p2.frameshift, False)
eq_(p2.ends_with_stop_codon, False)
eq_(p2.mutant_amino_acids, "")
eq_(p2.num_mutant_amino_acids, 0)
def make_dummy_protein_sequence(
n_supporting_variant_reads,
n_supporting_variant_sequences,
n_supporting_reference_transcripts,
n_total_variant_sequences=None,
n_total_variant_reads=None,
n_total_reference_transcripts=None,
amino_acids="MKHW", # ATG=M|AAA=K|CAC=H|TGG=W
cdna_sequence="CCCATGAAACACTGGTAG",
variant_cdna_interval_start=8, # assuming variant was AAC>AAA
variant_cdna_interval_end=9,
variant_aa_interval_start=1,
variant_aa_interval_end=2,
num_mismatches=1):
"""
Creates ProteinSequence object with None filled in for most fields
"""
if n_total_variant_reads is None:
n_total_variant_reads = n_supporting_variant_reads
if n_total_variant_sequences is None:
n_total_variant_sequences = n_supporting_variant_sequences
if n_total_reference_transcripts is None:
n_total_reference_transcripts = n_total_reference_transcripts
assert n_supporting_variant_sequences <= n_supporting_variant_reads
assert n_supporting_variant_sequences <= n_total_variant_sequences
assert n_supporting_reference_transcripts <= n_total_reference_transcripts
n_translations = n_total_reference_transcripts * n_total_variant_sequences
translation = make_dummy_translation(
amino_acids=amino_acids,
cdna_sequence=cdna_sequence,
offset_to_first_complete_codon=3,
variant_cdna_interval_start=variant_cdna_interval_start, # assuming variant was AAC>AAA
variant_cdna_interval_end=variant_cdna_interval_end,
variant_aa_interval_start=variant_aa_interval_start,
variant_aa_interval_end=variant_aa_interval_end,
num_mismatches=num_mismatches,
n_variant_reads=n_total_variant_reads)
return ProteinSequence(
translations=[translation] * n_translations)
def test_protein_sequence_deletion():
# testing that we got the correct properties in the case
# where "SIINFEKL" was mutated into "SIINFEL" by deletion of one amino acid
p = ProteinSequence(amino_acids="SIINFEL",
contains_mutation=True,
mutation_start_idx=len("SIINFE"),
mutation_end_idx=len("SIINFE"),
ends_with_stop_codon=True,
frameshift=False,
translations=[])
eq_(p.amino_acids, "SIINFEL")
eq_(len(p), 7)
eq_(p.num_mutant_amino_acids, 0)
eq_(p.mutant_amino_acids, "")
eq_(p.contains_mutation, True)
eq_(p.contains_deletion, True)
eq_(p.frameshift, False)
eq_(p.ends_with_stop_codon, True)