def getGeneticChanges(self, abspos, variants):
""" Given the possible variants at chromosomal position abspos, determine
any "interesting" genetic changes induced by each variant. That would
be any changes modifying a splice-site or interefering with the promoter
region
"""
import getGenomeSpan
### Stitch together the MRNA Sequence ...
sequence = []
changes = []
relpos = self.getRelpos(abspos)
sequence = getGenomeSpan.getGenomeSpan(self.chrom, self.txStart, self.txEnd)
sequence2 = list(sequence) ### This makes a copy
### For each variant specified, determine if there is an amino
### acid change in the protein when changing from the reference
### base at that position, to this "mutant" base
return []
if relpos < 0:
return 'Promoter?'
refBase = sequence[int(relpos)]
for variant in variants:
### Don't bother calculating a matching base, or deletions
if variant in (refBase, '-'):
continue
sequence2[positionInSequence] = variant
resp.append('NUCLEOTIDE=%s/%s' % (sequence[positionInSequence], sequence2[positionInSequence]))
sequence = ''.join(sequence)
sequence2 = ''.join(sequence2)
return changes
def getGeneticChanges(self, abspos, variants):
""" Given the possible variants at chromosomal position abspos, determine
any "interesting" genetic changes induced by each variant. That would
be any changes modifying a splice-site or interefering with the promoter
region
"""
import getGenomeSpan
### Stitch together the MRNA Sequence ...
sequence = []
changes = []
relpos = self.getRelpos(abspos)
sequence = getGenomeSpan.getGenomeSpan(self.chrom, self.txStart,
self.txEnd)
sequence2 = list(sequence) ### This makes a copy
### For each variant specified, determine if there is an amino
### acid change in the protein when changing from the reference
### base at that position, to this "mutant" base
return []
if relpos < 0:
return 'Promoter?'
refBase = sequence[int(relpos)]
for variant in variants:
### Don't bother calculating a matching base, or deletions
if variant in (refBase, '-'):
continue
sequence2[positionInSequence] = variant
resp.append(
'NUCLEOTIDE=%s/%s' %
(sequence[positionInSequence], sequence2[positionInSequence]))
sequence = ''.join(sequence)
sequence2 = ''.join(sequence2)
return changes
def getProteinChanges(self, abspos, variants):
"""
"""
import getGenomeSpan
### Stitch together the MRNA Sequence ...
sequence = []
variantPositions = []
for exonStart, exonEnd in self.exons:
### Forget exons that are entirely before cdsStart
### or after cdsEnd
if exonStart <= self.cdsStart and exonEnd <= self.cdsStart:
continue
elif exonStart >= self.cdsEnd and exonEnd >= self.cdsEnd:
continue
### Otherwise, add a CDS feature for the part of the
### exon contained in the range cdsStart-cdsEnd
fStart = max(exonStart, self.cdsStart)
fEnd = min(exonEnd, self.cdsEnd)
exon = getGenomeSpan.getGenomeSpan(self.chrom, fStart, fEnd)
sequence.append(exon)
### Determine the relative position within the mrna sequence
if fStart <= abspos <= fEnd:
variantPositions.append(len(sequence) - (fEnd - abspos))
changes = []
warnings = Set()
for positionInSequence in variantPositions:
sequence = list(''.join(sequence))
### For each variant specified, determine if there is an amino
### acid change in the protein when changing from the reference
### base at that position, to this "mutant" base
refBase = sequence[positionInSequence]
for variant in variants:
sequence2 = list(sequence)
change = {}
change['REF_BASE'] = refBase
### Don't bother calculating a matching base, or deletions
if variant in (refBase, '-'):
continue
sequence2[positionInSequence] = variant
change['VARIANT_BASE'] = sequence2[positionInSequence]
sequence = ''.join(sequence)
sequence2 = ''.join(sequence2)
mrna = Seq(sequence, alphabet)
mrna2 = Seq(sequence2, alphabet)
if self.strand == '-':
mrna = mrna.reverse_complement()
mrna2 = mrna2.reverse_complement()
protein = translator.translate(mrna)
protein2 = translator.translate(mrna2)
aminopos = positionInSequence/3
change['AMINOPOS'] = aminopos+1
### These two warning flags are here temporarily so we can get an idea
### of how often this happens. We've noticed sequences of stitched-together
### coding sequences that are not divisible by three, so when they are
### translated into proteins, the remaining bases get truncated. This is
### a general problem in that I don't know how best to handle that, and
### whether it's supposed to happen at all. It is a more specific problem
### when the SNP of interest is actually one of the truncated bases!
if len(mrna) % 3 != 0:
warnings.add('MRNA_LENGTH_NOT_3_MULTIPLE')
if len(protein) <= aminopos:
warnings.add('PROTEIN_LENGTH_LESS_THAN_AMINOPOS')
change['PROTEIN_LENGTH'] = len(protein)
continue
refAmino = protein[aminopos]
varAmino = protein2[aminopos]
if varAmino != refAmino:
### Some of these are rather verbose and are only useful for debugging
#change['HEAD'] = ''.join([str(i)[-2:].ljust(3) for i in range(1, len(protein)+1)])
#change['PRO1'] = ' '.join(protein.tostring())
#change['PRO2'] = ' '.join(protein2.tostring())
#change['SEQ1'] = mrna.tostring()
#change['SEQ2'] = mrna2.tostring()
change['VARIANT_POS'] = positionInSequence
change['CODING_LENGTH'] = len(sequence)
change['PROTEIN_LENGTH'] = len(protein)
change['REF_AMINO'] = refAmino
change['VAR_AMINO'] = varAmino
changes.append(change)
return changes, warnings
def getProteinChanges(self, abspos, variants):
"""
"""
import getGenomeSpan
### Stitch together the MRNA Sequence ...
sequence = []
variantPositions = []
for exonStart, exonEnd in self.exons:
### Forget exons that are entirely before cdsStart
### or after cdsEnd
if exonStart <= self.cdsStart and exonEnd <= self.cdsStart:
continue
elif exonStart >= self.cdsEnd and exonEnd >= self.cdsEnd:
continue
### Otherwise, add a CDS feature for the part of the
### exon contained in the range cdsStart-cdsEnd
fStart = max(exonStart, self.cdsStart)
fEnd = min(exonEnd, self.cdsEnd)
exon = getGenomeSpan.getGenomeSpan(self.chrom, fStart, fEnd)
sequence.append(exon)
### Determine the relative position within the mrna sequence
if fStart <= abspos <= fEnd:
variantPositions.append(len(sequence) - (fEnd - abspos))
changes = []
warnings = Set()
for positionInSequence in variantPositions:
sequence = list(''.join(sequence))
### For each variant specified, determine if there is an amino
### acid change in the protein when changing from the reference
### base at that position, to this "mutant" base
refBase = sequence[positionInSequence]
for variant in variants:
sequence2 = list(sequence)
change = {}
change['REF_BASE'] = refBase
### Don't bother calculating a matching base, or deletions
if variant in (refBase, '-'):
continue
sequence2[positionInSequence] = variant
change['VARIANT_BASE'] = sequence2[positionInSequence]
sequence = ''.join(sequence)
sequence2 = ''.join(sequence2)
mrna = Seq(sequence, alphabet)
mrna2 = Seq(sequence2, alphabet)
if self.strand == '-':
mrna = mrna.reverse_complement()
mrna2 = mrna2.reverse_complement()
protein = translator.translate(mrna)
protein2 = translator.translate(mrna2)
aminopos = positionInSequence / 3
change['AMINOPOS'] = aminopos + 1
### These two warning flags are here temporarily so we can get an idea
### of how often this happens. We've noticed sequences of stitched-together
### coding sequences that are not divisible by three, so when they are
### translated into proteins, the remaining bases get truncated. This is
### a general problem in that I don't know how best to handle that, and
### whether it's supposed to happen at all. It is a more specific problem
### when the SNP of interest is actually one of the truncated bases!
if len(mrna) % 3 != 0:
warnings.add('MRNA_LENGTH_NOT_3_MULTIPLE')
if len(protein) <= aminopos:
warnings.add('PROTEIN_LENGTH_LESS_THAN_AMINOPOS')
change['PROTEIN_LENGTH'] = len(protein)
continue
refAmino = protein[aminopos]
varAmino = protein2[aminopos]
if varAmino != refAmino:
### Some of these are rather verbose and are only useful for debugging
#change['HEAD'] = ''.join([str(i)[-2:].ljust(3) for i in range(1, len(protein)+1)])
#change['PRO1'] = ' '.join(protein.tostring())
#change['PRO2'] = ' '.join(protein2.tostring())
#change['SEQ1'] = mrna.tostring()
#change['SEQ2'] = mrna2.tostring()
change['VARIANT_POS'] = positionInSequence
change['CODING_LENGTH'] = len(sequence)
change['PROTEIN_LENGTH'] = len(protein)
change['REF_AMINO'] = refAmino
change['VAR_AMINO'] = varAmino
changes.append(change)
return changes, warnings
