def test_reference_coding_sequence_key_around_TP53_201_variant():
# TP53-201 is an isoform of TP53 which seems to lack untranslated
# regions so the sequence is:
# First exon: chr17 7,676,594 - 7,676,521
# ATG|GAG|GAG|CCG|CAG|TCA|GAT...
# -M-|-E-|-E-|-P-|-Q-|-S-|-D-
# we're assuming a variant
# chr17. 7,676,591 C>T which changes GAG (E) > AAG (K)
variant = Variant("chr17", 7676591, "C", "T", "GRCh38")
# TP53-201
transcript = variant.ensembl.transcripts_by_name("TP53-201")[0]
effect = variant.effect_on_transcript(transcript)
eq_(effect.__class__.__name__, "Substitution")
eq_(effect.aa_ref, "E")
eq_(effect.aa_alt, "K")
expected = ReferenceCodingSequenceKey(strand="-",
sequence_before_variant_locus="ATG",
sequence_at_variant_locus="G",
sequence_after_variant_locus="AGG",
offset_to_first_complete_codon=0,
contains_start_codon=True,
overlaps_start_codon=True,
contains_five_prime_utr=False,
amino_acids_before_variant="M")
reference_coding_sequence_key = ReferenceCodingSequenceKey.from_variant_and_transcript(
variant=variant, transcript=transcript, context_size=3)
eq_(expected, reference_coding_sequence_key)
def test_reference_coding_sequence_key_around_TP53_201_variant():
# TP53-201 is an isoform of TP53 which seems to lack untranslated
# regions so the sequence is:
# First exon: chr17 7,676,594 - 7,676,521
# ATG|GAG|GAG|CCG|CAG|TCA|GAT...
# -M-|-E-|-E-|-P-|-Q-|-S-|-D-
# we're assuming a variant
# chr17. 7,676,591 C>T which changes GAG (E) > AAG (K)
variant = Variant("chr17", 7676591, "C", "T", "GRCh38")
# TP53-201
transcript = variant.ensembl.transcripts_by_name("TP53-201")[0]
effect = variant.effect_on_transcript(transcript)
eq_(effect.__class__.__name__, "Substitution")
eq_(effect.aa_ref, "E")
eq_(effect.aa_alt, "K")
expected = ReferenceCodingSequenceKey(
strand="-",
sequence_before_variant_locus="ATG",
sequence_at_variant_locus="G",
sequence_after_variant_locus="AGG",
offset_to_first_complete_codon=0,
contains_start_codon=True,
overlaps_start_codon=True,
contains_five_prime_utr=False,
amino_acids_before_variant="M",
)
reference_coding_sequence_key = ReferenceCodingSequenceKey.from_variant_and_transcript(
variant=variant, transcript=transcript, context_size=3
)
eq_(expected, reference_coding_sequence_key)
def make_inputs_for_tp53_201_variant(
cdna_prefix="ATG",
cdna_suffix="AGGAGCCGCAGTCAGAT",
n_bad_nucleotides_at_start=0,
mismatches_before_variant=0,
mismatches_after_variant=14, # the read is that much longer than the reference (17 vs 3)
reference_context_size=3):
"""
Parameters
----------
cdna_prefix : str
Transcript nucleotides before the variant that we're pretending
got detected from RNA-seq reads.
cdna_suffix : str
Transcript nucleotides after the variant that we're pretending
got detected from RNA-seq reads.
n_bad_nucleotides_at_start : int
Number of nucleotides we expect to get trimmed from the
beginning of the variant sequence while matching to a reference context.
mismatches_before_variant : int
Expected number of nucleotide mismatches in the result before
the variant locus.
reference_context_size : int
Number of nucleotides before the variant locus to try matching
against a reference transcript.
"""
# TP53-201 is an isoform of TP53 which seems to lack untranslated
# regions so the sequence is:
# First exon: chr17 7,676,594 - 7,676,521
# ATG|GAG|GAG|CCG|CAG|TCA|GAT...
# -M-|-E-|-E-|-P-|-Q-|-S-|-D-
# we're assuming a variant
# chr17. 7,676,591 C>T which changes GAG (E) > AAG (K)
variant = Variant("chr17", 7676591, "C", "T", "GRCh38")
# TP53-201
transcript = variant.ensembl.transcripts_by_name("TP53-201")[0]
effect = variant.effect_on_transcript(transcript)
eq_(effect.__class__.__name__, "Substitution")
eq_(effect.aa_ref, "E")
eq_(effect.aa_alt, "K")
cdna_alt = "A"
# genomic DNA is the reverse complement of the cDNA
# for TP53-001 since it's on the negative strand
gdna_prefix = reverse_complement_dna(cdna_suffix)
gdna_alt = reverse_complement_dna(cdna_alt)
gdna_suffix = reverse_complement_dna(cdna_prefix)
# variant sequence supported by two reads
# one fully spanning the variant sequence
# and another missing the last nucleotide
fully_overlapping_read = AlleleRead(prefix=gdna_prefix,
allele=gdna_alt,
suffix=gdna_suffix,
name="full-overlap")
# testing the prefix and allele to make sure they have the expected
# TP53-201 sequence but the suffix might change depending on what's
# passed in as cdna_prefix
if cdna_suffix == "AGGAGCCGCAGTCAGAT":
eq_(fully_overlapping_read.prefix, "ATCTGACTGCGGCTCCT")
eq_(fully_overlapping_read.allele, "T")
partially_overlapping_read = AlleleRead(prefix=gdna_prefix,
allele=gdna_alt,
suffix=gdna_suffix[:-1],
name="partial-overlap")
if cdna_suffix == "AGGAGCCGCAGTCAGAT":
eq_(partially_overlapping_read.prefix, "ATCTGACTGCGGCTCCT")
eq_(partially_overlapping_read.allele, "T")
variant_sequence = VariantSequence(
prefix=gdna_prefix,
alt=gdna_alt,
suffix=gdna_suffix,
reads=[fully_overlapping_read, partially_overlapping_read])
assert isinstance(variant_sequence, VariantSequence)
prefix_length = len(cdna_prefix) - n_bad_nucleotides_at_start
reference_coding_sequence_key = ReferenceCodingSequenceKey.from_variant_and_transcript(
variant=variant,
transcript=transcript,
context_size=reference_context_size)
assert isinstance(reference_coding_sequence_key,
ReferenceCodingSequenceKey)
reference_context = ReferenceContext.from_reference_coding_sequence_key(
key=reference_coding_sequence_key,
variant=variant,
transcripts=[transcript])
assert isinstance(reference_context, ReferenceContext)
expected = VariantSequenceInReadingFrame(
cdna_sequence=cdna_prefix[-prefix_length:] + cdna_alt + cdna_suffix,
offset_to_first_complete_codon=prefix_length % 3,
variant_cdna_interval_start=prefix_length,
variant_cdna_interval_end=prefix_length + 1,
reference_cdna_sequence_before_variant="ATG"[-prefix_length:],
reference_cdna_sequence_after_variant=
"AGGAGCCGCAGTCAGAT"[:reference_context_size],
number_mismatches_before_variant=mismatches_before_variant,
number_mismatches_after_variant=mismatches_after_variant)
assert isinstance(expected, VariantSequenceInReadingFrame)
return variant_sequence, reference_context, expected
def make_inputs_for_tp53_201_variant(
cdna_prefix="ATG",
n_bad_nucleotides_at_start=0,
mismatches=0,
reference_context_size=3):
"""
Parameters
----------
cdna_prefix : str
Transcript nucleotides before the variant that we're pretending
got detected from RNA-seq reads.
n_bad_nucleotides_at_start : int
Number of nucleotides we expect to get trimmed from the
beginning of the variant sequence while matching to a reference context.
mismatches : int
Expected number of nucleotide mismatches in the result
reference_context_size : int
Number of nucleotides before the variant locus to try matching
against a reference transcript.
"""
# TP53-201 is an isoform of TP53 which seems to lack untranslated
# regions so the sequence is:
# First exon: chr17 7,676,594 - 7,676,521
# ATG|GAG|GAG|CCG|CAG|TCA|GAT...
# -M-|-E-|-E-|-P-|-Q-|-S-|-D-
# we're assuming a variant
# chr17. 7,676,591 C>T which changes GAG (E) > AAG (K)
variant = Variant("chr17", 7676591, "C", "T", "GRCh38")
# TP53-201
transcript = variant.ensembl.transcripts_by_name("TP53-201")[0]
effect = variant.effect_on_transcript(transcript)
eq_(effect.__class__.__name__, "Substitution")
eq_(effect.aa_ref, "E")
eq_(effect.aa_alt, "K")
cdna_alt = "A"
cdna_suffix = "AGGAGCCGCAGTCAGAT"
# genomic DNA is the reverse complement of the cDNA
# for TP53-001 since it's on the negative strand
gdna_prefix = reverse_complement_dna(cdna_suffix)
gdna_alt = reverse_complement_dna(cdna_alt)
gdna_suffix = reverse_complement_dna(cdna_prefix)
# variant sequence supported by two reads
# one fully spanning the variant sequence
# and another missing the last nucleotide
fully_overlapping_read = AlleleRead(
prefix=gdna_prefix,
allele=gdna_alt,
suffix=gdna_suffix,
name="full-overlap")
# testing the prefix and allele to make sure they have the expected
# TP53-201 sequence but the suffix might change depending on what's
# passed in as cdna_prefix
eq_(fully_overlapping_read.prefix, "ATCTGACTGCGGCTCCT")
eq_(fully_overlapping_read.allele, "T")
partially_overlapping_read = AlleleRead(
prefix=gdna_prefix,
allele=gdna_alt,
suffix=gdna_suffix[:-1],
name="partial-overlap")
eq_(partially_overlapping_read.prefix, "ATCTGACTGCGGCTCCT")
eq_(partially_overlapping_read.allele, "T")
variant_sequence = VariantSequence(
prefix=gdna_prefix,
alt=gdna_alt,
suffix=gdna_suffix,
reads=[fully_overlapping_read, partially_overlapping_read])
assert isinstance(variant_sequence, VariantSequence)
prefix_length = len(cdna_prefix) - n_bad_nucleotides_at_start
reference_coding_sequence_key = ReferenceCodingSequenceKey.from_variant_and_transcript(
variant=variant,
transcript=transcript,
context_size=reference_context_size)
assert isinstance(reference_coding_sequence_key, ReferenceCodingSequenceKey)
reference_context = ReferenceContext.from_reference_coding_sequence_key(
key=reference_coding_sequence_key,
variant=variant,
transcripts=[transcript])
assert isinstance(reference_context, ReferenceContext)
expected = VariantSequenceInReadingFrame(
cdna_sequence=cdna_prefix[-prefix_length:] + cdna_alt + cdna_suffix,
offset_to_first_complete_codon=prefix_length % 3,
variant_cdna_interval_start=prefix_length,
variant_cdna_interval_end=prefix_length + 1,
reference_cdna_sequence_before_variant="ATG"[-prefix_length:],
number_mismatches=mismatches)
assert isinstance(expected, VariantSequenceInReadingFrame)
return variant_sequence, reference_context, expected
