def removeRestrictionSites(self, sourceSeq, optimizedSeq, restrictionSites):
'''
get the best sequence that does not contain any restriction sites
by re-randomizing until no restriction site remains
'''
codons = re.findall('...', optimizedSeq)
remainder = SeqUtils.getRemainderSuffix(optimizedSeq)
restrictionSites = SeqUtils.expandAmbiguousMult(restrictionSites)
restrictionLocations = SeqUtils.searchSubseqs(optimizedSeq, restrictionSites)
restrictionCodons = SeqUtils.getCodonsForRanges(restrictionLocations)
# try to re-randomize a finite amount of times
ITERMAX = 10000
iteration = 0
while iteration < ITERMAX:
for i in restrictionCodons:
codons[i] = self.getRandomOptimizedCodon(codons[i])
tSeq = "".join(codons) + remainder
tRL = SeqUtils.searchSubseqs(tSeq, restrictionSites)
tRC = SeqUtils.getCodonsForRanges(tRL)
if not tRC:
return tSeq
iteration += 1
return None
def removeRestrictionSites(self, sourceSeq, optimizedSeq, restrictionSites):
'''
get the best sequence that does not contain any restriction sites
by substituting codons in the optimizedSeq in a shortest-paths manner
'''
checkedSeqs = set()
worklist = list()
restrictionSites = SeqUtils.expandAmbiguousMult(restrictionSites)
heapq.heappush(worklist, (self.scoreSequence(sourceSeq, optimizedSeq), optimizedSeq))
while len(worklist) > 0:
tScore, tSeq = heapq.heappop(worklist)
restrictionLocations = SeqUtils.searchSubseqs(tSeq, restrictionSites)
restrictionCodons = SeqUtils.getCodonsForRanges(restrictionLocations)
if not restrictionCodons:
return tSeq
else:
checkedSeqs.add(tSeq)
possibleChanges = self.getPossibleOneStepChanges(sourceSeq, tSeq, restrictionCodons)
if possibleChanges:
for tNewSeq in possibleChanges:
if not tNewSeq in checkedSeqs:
heapq.heappush(worklist, (self.scoreSequence(sourceSeq, tNewSeq), tNewSeq))
return None
def SequenceToPrint(self, seq, restrictionSites, source=True):
'''
return tuples of the form (codon, usage, isRestrictionSite) for the sequence
'''
codons = re.findall('...', seq)
remainder = SeqUtils.getRemainderSuffix(seq)
result = list()
restrictionSites = SeqUtils.expandAmbiguousMult(restrictionSites)
restrictionLocations = SeqUtils.searchSubseqs(seq, restrictionSites)
restrictionCodons = SeqUtils.getCodonsForRanges(restrictionLocations)
for i in range(len(codons)):
usage = self.sourceCU.getCodonRelativeUsage(codons[i]) if source else self.targetCU.getCodonRelativeUsage(codons[i])
result.append((codons[i], usage, i in restrictionCodons))
if remainder:
result.append((remainder, None, None))
return result