def _map(self, from_pos, to_pos, pos, end, strict_bounds):
"""Map position between aligned segments
Positions in this function are 0-based, base-counting.
"""
if strict_bounds and (pos < 0 or pos > from_pos[-1]):
raise HGVSInvalidIntervalError("Position is beyond the bounds of transcript record")
# find aligned segment to use as basis for mapping
# okay for pos to be before first element or after last
for pos_i in range(len(self.cigar_op)):
if pos < from_pos[pos_i+1]:
break
if self.cigar_op[pos_i] in "=MX":
mapped_pos = to_pos[pos_i] + (pos - from_pos[pos_i])
mapped_pos_offset = 0
elif self.cigar_op[pos_i] in "DI":
mapped_pos = to_pos[pos_i]
if end == "start":
mapped_pos -= 1
mapped_pos_offset = 0
elif self.cigar_op[pos_i] == "N":
if pos - from_pos[pos_i] + 1 <= from_pos[pos_i + 1] - pos:
mapped_pos = to_pos[pos_i] - 1
mapped_pos_offset = pos - from_pos[pos_i] + 1
else:
mapped_pos = to_pos[pos_i]
mapped_pos_offset = -(from_pos[pos_i + 1] - pos)
return mapped_pos, mapped_pos_offset, self.cigar_op[pos_i]
def _map(self, from_pos, to_pos, pos, base):
"""Map position between aligned sequences
Positions in this function are 0-based.
"""
pos_i = -1
while pos_i < len(self.cigar_op) and pos >= from_pos[pos_i + 1]:
pos_i += 1
if pos_i == -1 or pos_i == len(self.cigar_op):
raise HGVSInvalidIntervalError(
"Position is beyond the bounds of transcript record")
if self.cigar_op[pos_i] in "=MX":
mapped_pos = to_pos[pos_i] + (pos - from_pos[pos_i])
mapped_pos_offset = 0
elif self.cigar_op[pos_i] in "DI":
if base == "start":
mapped_pos = to_pos[pos_i] - 1
elif base == "end":
mapped_pos = to_pos[pos_i]
mapped_pos_offset = 0
elif self.cigar_op[pos_i] == "N":
if pos - from_pos[pos_i] + 1 <= from_pos[pos_i + 1] - pos:
mapped_pos = to_pos[pos_i] - 1
mapped_pos_offset = pos - from_pos[pos_i] + 1
else:
mapped_pos = to_pos[pos_i]
mapped_pos_offset = -(from_pos[pos_i + 1] - pos)
return mapped_pos, mapped_pos_offset, self.cigar_op[pos_i]
def _map(from_ivs, to_ivs, from_start_i, from_end_i, max_extent):
def iv_map(from_ivs, to_ivs, from_start_i, from_end_i, max_extent):
"""returns the <start,end> intervals indexes in which from_start_i and from_end_i occur"""
# first look for 0-width interval that matches
seil = [i for i, iv in enumerate(from_ivs) if iv.start_i == from_start_i and iv.end_i == from_end_i]
if len(seil) > 0:
si = ei = seil[0]
else:
sil = [i for i, iv in enumerate(from_ivs) if iv.start_i <= from_start_i <= iv.end_i]
eil = [i for i, iv in enumerate(from_ivs) if iv.start_i <= from_end_i <= iv.end_i]
if len(sil) == 0 or len(eil) == 0:
raise HGVSInvalidIntervalError("start or end or both are beyond the bounds of transcript record")
si, ei = (sil[0], eil[-1]) if max_extent else (sil[-1], eil[0])
return si, ei
def clip_to_iv(iv, pos):
return max(iv.start_i, min(iv.end_i, pos))
assert from_start_i <= from_end_i, "expected from_start_i <= from_end_i"
try:
si, ei = iv_map(from_ivs, to_ivs, from_start_i, from_end_i, max_extent)
except ValueError:
raise HGVSInvalidIntervalError("start_i,end_i interval out of bounds")
to_start_i = clip_to_iv(to_ivs[si], to_ivs[si].start_i + (from_start_i - from_ivs[si].start_i))
to_end_i = clip_to_iv(to_ivs[ei], to_ivs[ei].end_i - (from_ivs[ei].end_i - from_end_i))
return to_start_i, to_end_i
def n_to_c(self, n_interval, strict_bounds=None):
"""convert a transcript cDNA (n.) interval to a transcript CDS (c.) interval"""
if strict_bounds is None:
strict_bounds = global_config.mapping.strict_bounds
if self.cds_start_i is None: # cds_start_i defined iff cds_end_i defined; see assertion above
raise HGVSUsageError(
"CDS is undefined for {self.tx_ac}; cannot map to c. coordinate (non-coding transcript?)"
.format(self=self))
if strict_bounds and (n_interval.start.base <= 0 or n_interval.end.base > self.tgt_len):
raise HGVSInvalidIntervalError(
"The given coordinate is outside the bounds of the reference sequence.")
def pos_n_to_c(pos):
if pos.base <= self.cds_start_i:
c = pos.base - self.cds_start_i - (1 if pos.base > 0 else 0)
c_datum = Datum.CDS_START
elif pos.base > self.cds_start_i and pos.base <= self.cds_end_i:
c = pos.base - self.cds_start_i
c_datum = Datum.CDS_START
else:
c = pos.base - self.cds_end_i
c_datum = Datum.CDS_END
return hgvs.location.BaseOffsetPosition(base=c, offset=pos.offset, datum=c_datum)
c_interval = hgvs.location.BaseOffsetInterval(start=pos_n_to_c(n_interval.start),
end=pos_n_to_c(n_interval.end),
uncertain=n_interval.uncertain)
return c_interval
def __init__(self, ref, tgt):
if not ((ref.len == tgt.len) or (ref.len == 0 and tgt.len != 0) or
(ref.len != 0 and tgt.len == 0)):
raise HGVSInvalidIntervalError(
"IntervalPair doesn't represent a match, insertion, or deletion"
)
self.ref = ref
self.tgt = tgt
def check_datum(self):
# check for valid combinations of start and end datums
if (self.start.datum, self.end.datum) not in [
(Datum.SEQ_START, Datum.SEQ_START),
(Datum.CDS_START, Datum.CDS_START),
(Datum.CDS_START, Datum.CDS_END),
(Datum.CDS_END, Datum.CDS_END),
]:
raise HGVSInvalidIntervalError("BaseOffsetInterval start datum and end datum are incompatible")
def iv_map(from_ivs, to_ivs, from_start_i, from_end_i, max_extent):
"""returns the <start,end> intervals indexes in which from_start_i and from_end_i occur"""
# first look for 0-width interval that matches
seil = [i for i, iv in enumerate(from_ivs) if iv.start_i == from_start_i and iv.end_i == from_end_i]
if len(seil) > 0:
si = ei = seil[0]
else:
sil = [i for i, iv in enumerate(from_ivs) if iv.start_i <= from_start_i <= iv.end_i]
eil = [i for i, iv in enumerate(from_ivs) if iv.start_i <= from_end_i <= iv.end_i]
if len(sil) == 0 or len(eil) == 0:
raise HGVSInvalidIntervalError("start or end or both are beyond the bounds of transcript record")
si, ei = (sil[0], eil[-1]) if max_extent else (sil[-1], eil[0])
return si, ei
def pos_c_to_n(pos):
if pos.datum == Datum.CDS_START:
n = pos.base + self.cds_start_i
if pos.base < 0: # correct for lack of c.0 coordinate
n += 1
elif pos.datum == Datum.CDS_END:
n = pos.base + self.cds_end_i
if n <= 0: # correct for lack of n.0 coordinate
n -= 1
if (n <= 0 or n > self.tgt_len):
if strict_bounds:
raise HGVSInvalidIntervalError(f"c.{pos} coordinate is out of bounds")
return hgvs.location.BaseOffsetPosition(base=n,
offset=pos.offset,
datum=Datum.SEQ_START)
def c_to_n(self, c_interval, strict_bounds=None):
"""convert a transcript CDS (c.) interval to a transcript cDNA (n.) interval"""
if strict_bounds is None:
strict_bounds = global_config.mapping.strict_bounds
if self.cds_start_i is None: # cds_start_i defined iff cds_end_i defined; see assertion above
raise HGVSUsageError(
"CDS is undefined for {self.tx_ac}; cannot map from c. coordinate (non-coding transcript?)"
.format(self=self))
# start
if c_interval.start.datum == Datum.CDS_START and c_interval.start.base < 0:
n_start = c_interval.start.base + self.cds_start_i + 1
elif c_interval.start.datum == Datum.CDS_START and c_interval.start.base > 0:
n_start = c_interval.start.base + self.cds_start_i
elif c_interval.start.datum == Datum.CDS_END:
n_start = c_interval.start.base + self.cds_end_i
# end
if c_interval.end.datum == Datum.CDS_START and c_interval.end.base < 0:
n_end = c_interval.end.base + self.cds_start_i + 1
elif c_interval.end.datum == Datum.CDS_START and c_interval.end.base > 0:
n_end = c_interval.end.base + self.cds_start_i
elif c_interval.end.datum == Datum.CDS_END:
n_end = c_interval.end.base + self.cds_end_i
if strict_bounds and (n_start <= 0 or n_end > self.tgt_len):
raise HGVSInvalidIntervalError(
"The given coordinate is outside the bounds of the reference sequence."
)
n_interval = hgvs.location.BaseOffsetInterval(
start=hgvs.location.BaseOffsetPosition(
base=n_start,
offset=c_interval.start.offset,
datum=Datum.SEQ_START),
end=hgvs.location.BaseOffsetPosition(base=n_end,
offset=c_interval.end.offset,
datum=Datum.SEQ_START),
uncertain=c_interval.uncertain)
return n_interval
def n_to_c(self, n_interval):
"""convert a transcript cDNA (n.) interval to a transcript CDS (c.) interval"""
if self.cds_start_i is None: # cds_start_i defined iff cds_end_i defined; see assertion above
raise HGVSUsageError(
"CDS is undefined for {self.tx_ac}; cannot map to c. coordinate (non-coding transcript?)"
.format(self=self))
if n_interval.start.base <= 0 or n_interval.end.base > self.tgt_len:
raise HGVSInvalidIntervalError(
"The given coordinate is outside the bounds of the reference sequence."
)
# start
if n_interval.start.base <= self.cds_start_i:
cs = n_interval.start.base - (self.cds_start_i + 1)
cs_datum = Datum.CDS_START
elif n_interval.start.base > self.cds_start_i and n_interval.start.base <= self.cds_end_i:
cs = n_interval.start.base - self.cds_start_i
cs_datum = Datum.CDS_START
else:
cs = n_interval.start.base - self.cds_end_i
cs_datum = Datum.CDS_END
# end
if n_interval.end.base <= self.cds_start_i:
ce = n_interval.end.base - (self.cds_start_i + 1)
ce_datum = Datum.CDS_START
elif n_interval.end.base > self.cds_start_i and n_interval.end.base <= self.cds_end_i:
ce = n_interval.end.base - self.cds_start_i
ce_datum = Datum.CDS_START
else:
ce = n_interval.end.base - self.cds_end_i
ce_datum = Datum.CDS_END
c_interval = hgvs.location.BaseOffsetInterval(
start=hgvs.location.BaseOffsetPosition(
base=cs, offset=n_interval.start.offset, datum=cs_datum),
end=hgvs.location.BaseOffsetPosition(base=ce,
offset=n_interval.end.offset,
datum=ce_datum),
uncertain=n_interval.uncertain)
return c_interval
def __init__(self, start_i, end_i):
if not (start_i <= end_i):
raise HGVSInvalidIntervalError(
"start_i must be less than or equal to end_i")
self.start_i = start_i
self.end_i = end_i