def _determine_codon_overlap(self, s, e, codon_tuple, variant_type):
if codon_tuple is None:
return False
if variant_type == VariantClassification.VT_INS:
is_codon_overlap = TranscriptProviderUtils.test_overlap(s, s, codon_tuple[0]+1, codon_tuple[1])
else:
is_codon_overlap = TranscriptProviderUtils.test_overlap(s, e, codon_tuple[0]+1, codon_tuple[1])
return is_codon_overlap
def _determine_codon_overlap(self, s, e, codon_tuple, variant_type):
if codon_tuple is None:
return False
if variant_type == VariantClassification.VT_INS:
is_codon_overlap = TranscriptProviderUtils.test_overlap(
s, s, codon_tuple[0] + 1, codon_tuple[1])
else:
is_codon_overlap = TranscriptProviderUtils.test_overlap(
s, e, codon_tuple[0] + 1, codon_tuple[1])
return is_codon_overlap
def _determine_if_splice_site_overlap(self,
start_genomic_space,
end_genomic_space,
tx,
variant_type,
dist=2):
"""
Overlap of start and stop codon (i.e. start of first exon and end of last exon -- stranded) will not be a
Splice_Site. This method will return is_splice_site_overlap of False
If overlap is detected, but the start or end is within dist bp, then this is a splice site.
start <= end
INS events only call splice site when they start in the splice site
:param start_genomic_space: int in genomic space
:param end_genomic_space: int in genomic space
:param tx: Transcript
:param variant_type:
:param dist:
:return is_splice_site_overlap, exon_i, is_right_overlap (Higher genomic position --> True)
"""
exons = tx.get_exons()
strand = tx.get_strand()
# If this is an insertion, we only want to count a splice site if it starts in the splice site regions
if variant_type == VariantClassification.VT_INS:
end_genomic_space = start_genomic_space
for i, exon in enumerate(exons):
is_internal_exon = (i > 0) and (i < (len(exons) - 1))
is_check_left = is_internal_exon or (strand == "-" and i == 0) or (
strand == "+" and i == (len(exons) - 1))
is_check_right = is_internal_exon or (
strand == "+" and i == 0) or (strand == "-"
and i == (len(exons) - 1))
if is_check_left:
splice_site_left = (exon[0] - dist + 1,
exon[0] + (dist - 1) + 1)
overlap_type_left = TranscriptProviderUtils.test_overlap(
start_genomic_space, end_genomic_space,
splice_site_left[0], splice_site_left[1])
if overlap_type_left:
return True, i, False
if is_check_right:
splice_site_right = (exon[1] - (dist - 1), exon[1] + dist)
overlap_type_right = TranscriptProviderUtils.test_overlap(
start_genomic_space, end_genomic_space,
splice_site_right[0], splice_site_right[1])
if overlap_type_right:
return True, i, True
return False, -1, None, False
def _is_matching(self, mut, tsv_record):
chrom = tsv_record[self.tsv_index["chrom"]]
startPos = tsv_record[self.tsv_index["start"]]
endPos = tsv_record[self.tsv_index["end"]]
build = "hg19"
if self.match_mode == "exact":
if "ref" in self.tsv_index and "alt" in self.tsv_index: # ref and alt information is present
ref = tsv_record[self.tsv_index["ref"]]
alt = tsv_record[self.tsv_index["alt"]]
if ref == "-" or alt == "-": # addresses Mutation Annotation Format based tsv records
# TODO: This looks risky to be calling the MutationData constructor directly
ds_mut = MutationData(chrom, startPos, endPos, ref, alt, build)
else: # addresses tsv records where the input isn't a Mutation Annotation Format file
ds_mut = MutUtils.initializeMutFromAttributes(chrom, startPos, endPos, ref, alt, build)
if mut.chr == ds_mut.chr and mut.ref_allele == ds_mut.ref_allele \
and mut.alt_allele == ds_mut.alt_allele and int(mut.start) == int(ds_mut.start) \
and int(mut.end) == int(ds_mut.end):
return True
else: # do not use ref and alt information
if mut.chr == chrom and int(mut.start) == int(startPos) and int(mut.end) == int(endPos):
return True
else:
return TranscriptProviderUtils.test_overlap(int(mut.start), int(mut.end), int(startPos), int(endPos))
return False
def _is_matching(self, mut, tsv_record):
chrom = tsv_record[self.tsv_index["chrom"]]
startPos = tsv_record[self.tsv_index["start"]]
endPos = tsv_record[self.tsv_index["end"]]
build = "hg19"
if self.match_mode == "exact":
if "ref" in self.tsv_index and "alt" in self.tsv_index: # ref and alt information is present
ref = tsv_record[self.tsv_index["ref"]]
alt = tsv_record[self.tsv_index["alt"]]
if ref == "-" or alt == "-": # addresses Mutation Annotation Format based tsv records
# TODO: This looks risky to be calling the MutationData constructor directly
ds_mut = MutationData(chrom, startPos, endPos, ref, alt,
build)
else: # addresses tsv records where the input isn't a Mutation Annotation Format file
ds_mut = MutUtils.initializeMutFromAttributes(
chrom, startPos, endPos, ref, alt, build)
if mut.chr == ds_mut.chr and mut.ref_allele == ds_mut.ref_allele \
and mut.alt_allele == ds_mut.alt_allele and int(mut.start) == int(ds_mut.start) \
and int(mut.end) == int(ds_mut.end):
return True
else: # do not use ref and alt information
if mut.chr == chrom and int(
mut.start) == int(startPos) and int(
mut.end) == int(endPos):
return True
else:
return TranscriptProviderUtils.test_overlap(
int(mut.start), int(mut.end), int(startPos), int(endPos))
return False
def _determine_if_splice_site_overlap(self, start_genomic_space, end_genomic_space, tx, variant_type, dist=2):
"""
Overlap of start and stop codon (i.e. start of first exon and end of last exon -- stranded) will not be a
Splice_Site. This method will return is_splice_site_overlap of False
If overlap is detected, but the start or end is within dist bp, then this is a splice site.
start <= end
INS events only call splice site when they start in the splice site
:param start_genomic_space: int in genomic space
:param end_genomic_space: int in genomic space
:param tx: Transcript
:param variant_type:
:param dist:
:return is_splice_site_overlap, exon_i, is_right_overlap (Higher genomic position --> True)
"""
exons = tx.get_exons()
strand = tx.get_strand()
# If this is an insertion, we only want to count a splice site if it starts in the splice site regions
if variant_type == VariantClassification.VT_INS:
end_genomic_space = start_genomic_space
for i,exon in enumerate(exons):
is_internal_exon = (i > 0) and (i < (len(exons)-1))
is_check_left = is_internal_exon or (strand == "-" and i == 0) or (strand == "+" and i == (len(exons)-1))
is_check_right = is_internal_exon or (strand == "+" and i == 0) or (strand == "-" and i == (len(exons)-1))
if is_check_left:
splice_site_left = (exon[0]-dist+1, exon[0]+(dist-1)+1)
overlap_type_left = TranscriptProviderUtils.test_overlap(start_genomic_space, end_genomic_space, splice_site_left[0], splice_site_left[1])
if overlap_type_left:
return True, i, False
if is_check_right:
splice_site_right = (exon[1]-(dist-1), exon[1] + dist)
overlap_type_right = TranscriptProviderUtils.test_overlap(start_genomic_space, end_genomic_space, splice_site_right[0], splice_site_right[1])
if overlap_type_right:
return True, i, True
return False, -1, None, False
def __get_overlapping_records(self, records, start, end, type):
if type == "gene":
st_key, en_key = "start", "end"
elif type == "transcript":
st_key, en_key = "footprint_start", "footprint_end"
out_records = list()
for r in records:
if TranscriptProviderUtils.test_overlap(start, end, r[st_key], r[en_key]):
out_records.append(r)
return out_records
def __get_overlapping_records(self, records, start, end, type):
if type == 'gene':
st_key, en_key = 'start', 'end'
elif type == 'transcript':
st_key, en_key = 'footprint_start', 'footprint_end'
out_records = list()
for r in records:
if TranscriptProviderUtils.test_overlap(start, end, r[st_key], r[en_key]):
out_records.append(r)
return out_records
def __get_overlapping_records(self, records, start, end, type):
if type == 'gene':
st_key, en_key = 'start', 'end'
elif type == 'transcript':
st_key, en_key = 'footprint_start', 'footprint_end'
out_records = list()
for r in records:
if TranscriptProviderUtils.test_overlap(start, end, r[st_key],
r[en_key]):
out_records.append(r)
return out_records
def _get_overlapping_transcript_records(self, records, start, end):
return [
r for r in records if TranscriptProviderUtils.test_overlap(
int(start), int(end), r.get_start(), r.get_end())
]
def _get_overlapping_transcript_records(self, records, start, end):
return [r for r in records if TranscriptProviderUtils.test_overlap(int(start), int(end), r.get_start(), r.get_end())]
