def findVoteScore(in1List, leftSeg, parameterRobot):
H0Score, H1Score = 0, 0
threshold = 10
perr = parameterRobot.p
print " len(in1List) ", len(in1List)
for i in range(len(in1List)):
# Determine F/B
read = in1List[i].longread
score, returnalignedSeq1, returnalignedSeq2, starti, startj, endi, endj = cleaner.SWAlignment(
read, leftSeg[0], parameterRobot)
reverseread = cleaner.reverseStrand(read)
scoreRev, returnalignedSeq1Rev, returnalignedSeq2Rev, startiRev, startjRev, endiRev, endjRev = cleaner.SWAlignment(
reverseread, leftSeg[0], parameterRobot)
if score > scoreRev:
forwardCk = True
scoreLeft0 = score
else:
forwardCk = False
scoreLeft0 = scoreRev
if forwardCk:
score, returnalignedSeq1, returnalignedSeq2, starti, startj, endi, endj = cleaner.SWAlignment(
read, leftSeg[1], parameterRobot)
scoreLeft1 = score
else:
score, returnalignedSeq1, returnalignedSeq2, starti, startj, endi, endj = cleaner.SWAlignment(
reverseread, leftSeg[1], parameterRobot)
scoreLeft1 = score
print "scoreLeft0, scoreLeft1 ", scoreLeft0, scoreLeft1
skipCounting = False
MAPScore = [0, 0]
# Filering reads
if startj > threshold or endj < len(leftSeg[0]) - threshold:
skipCounting = True
# MAP decision rule
if forwardCk:
for i in range(2):
#tempScore, returnalignedSeq1, returnalignedSeq2 , starti, startj , endi, endj, numPath= cleaner.SWAlignmentFixRefMP(leftSeg[i],read, parameterRobot)
#lengthSeed = len(leftSeg[i])
#countConfirm, countDel, countIns, countSub = cleaner.countEdits(returnalignedSeq2, returnalignedSeq1) # care for the 1,2 and its fcn
#confirmNoDelete,confirmNoInsert = countConfirm , lengthSeed - countIns
#MAPScore[i] = math.log(1-perr)*confirmNoDelete + math.log(1-perr)*confirmNoInsert + math.log(perr)*countDel + math.log(perr*perr/3)*countSub + math.log(perr/3)*countIns + math.log(numPath)
tempScore, returnalignedSeq1, returnalignedSeq2, starti, startj, endi, endj, numPath = cleaner.SWAlignmentFixRefMPQS(
leftSeg[i], read, parameterRobot)
MAPScore[i] = tempScore
else:
for i in range(2):
#tempScore, returnalignedSeq1, returnalignedSeq2 , starti, startj , endi, endj, numPath= cleaner.SWAlignmentFixRefMP(leftSeg[i],reverseread, parameterRobot)
#lengthSeed = len(leftSeg[i])
#countConfirm, countDel, countIns, countSub = cleaner.countEdits(returnalignedSeq2, returnalignedSeq1) # care for the 1,2 and its fcn
#confirmNoDelete,confirmNoInsert = countConfirm , lengthSeed - countIns
#MAPScore[i] = math.log(1-perr)*confirmNoDelete + math.log(1-perr)*confirmNoInsert + math.log(perr)*countDel + math.log(perr*perr/3)*countSub + math.log(perr/3)*countIns + math.log(numPath)
tempScore, returnalignedSeq1, returnalignedSeq2, starti, startj, endi, endj, numPath = cleaner.SWAlignmentFixRefMPQS(
leftSeg[i], reverseread, parameterRobot)
MAPScore[i] = tempScore
if not skipCounting:
H0Score += MAPScore[0]
H1Score += MAPScore[1]
return H0Score, H1Score
def linkSegments(inSeg, middleSeg, outSeg, parameterRobot):
print "\nNew Contig"
middleRead = middleSeg.longread
revmiddleRead = cleaner.reverseStrand(middleRead)
extendedRead = []
score, scoreRev = -1, -1
inIndex = 0
startjTmp = -1
while ((max(score, scoreRev) < 30 or startjTmp > 10)
and inIndex < len(inSeg)):
inRead = inSeg[inIndex].longread
revinRead = cleaner.reverseStrand(inRead)
score, returnalignedSeq1, returnalignedSeq2, starti, startj, endi, endj = cleaner.SWAlignment(
inRead, middleRead, parameterRobot)
scoreRev, returnalignedSeq1Rev, returnalignedSeq2Rev, startiRev, startjRev, endiRev, endjRev = cleaner.SWAlignment(
revinRead, middleRead, parameterRobot)
if score > scoreRev:
startjTmp = startj
print "starti, startj , endi, endj", starti, startj, endi, endj
else:
startjTmp = startjRev
print "startiRev, startjRev , endiRev, endjRev", startiRev, startjRev, endiRev, endjRev
print "startjTmp", startjTmp
print "score, scoreRev,inIndex ", score, scoreRev, inIndex
inIndex += 1
if score > scoreRev:
for eachbase in inRead[0:starti]:
extendedRead.append(eachbase)
print "len(extendedRead)", len(extendedRead)
for eachbase in middleRead:
extendedRead.append(eachbase)
print "len(extendedRead)", len(extendedRead)
else:
for eachbase in revinRead[0:startiRev]:
extendedRead.append(eachbase)
print "len(extendedRead)", len(extendedRead)
for eachbase in middleRead:
extendedRead.append(eachbase)
print "len(extendedRead)", len(extendedRead)
score, scoreRev = -1, -1
# hack
outIndex = 0
endiTmp = 1000
outRead, revoutRead = [], []
while ((max(score, scoreRev) < 30 or endiTmp < len(middleRead) - 10)
and outIndex < len(outSeg)):
outRead = outSeg[outIndex].longread
revoutRead = cleaner.reverseStrand(outRead)
score, returnalignedSeq1, returnalignedSeq2, starti, startj, endi, endj = cleaner.SWAlignment(
middleRead, outRead, parameterRobot)
scoreRev, returnalignedSeq1Rev, returnalignedSeq2Rev, startiRev, startjRev, endiRev, endjRev = cleaner.SWAlignment(
middleRead, revoutRead, parameterRobot)
if score > scoreRev:
endiTmp = endi
print "starti, startj , endi, endj", starti, startj, endi, endj
else:
endiTmp = endiRev
print " startiRev, startjRev , endiRev, endjRev ", startiRev, startjRev, endiRev, endjRev
print "score, scoreRev, outIndex", score, scoreRev, outIndex
outIndex += 1
correctionTerm = len(middleRead) - endiTmp
if score > scoreRev:
for eachbase in outRead[endj + correctionTerm:len(outRead)]:
extendedRead.append(eachbase)
print "len(extendedRead) forw", len(extendedRead)
else:
for eachbase in revoutRead[endjRev + correctionTerm:len(revoutRead)]:
extendedRead.append(eachbase)
print "len(extendedRead) rev", len(extendedRead)
return extendedRead
def decideMiddlePiece(commonList, leftSeg, rightSeg, parameterRobot):
middleMap = []
threshold = 10
perr = parameterRobot.p
H0, H1 = 0, 0
for i in range(len(commonList)):
read = commonList[i].longread
score, returnalignedSeq1, returnalignedSeq2, starti, startj, endi, endj = cleaner.SWAlignment(
read, leftSeg[0], parameterRobot)
readRev = cleaner.reverseStrand(read)
scoreRev, returnalignedSeq1Rev, returnalignedSeq2Rev, startiRev, startjRev, endiRev, endjRev = cleaner.SWAlignment(
readRev, leftSeg[0], parameterRobot)
if score > scoreRev:
forwardCk = True
scoreL0 = score
else:
forwardCk = False
scoreL0 = scoreRev
if forwardCk:
tmpread = read
else:
tmpread = readRev
scoreL1, returnalignedSeq1, returnalignedSeq2, starti, startj, endi, endj = cleaner.SWAlignment(
tmpread, leftSeg[1], parameterRobot)
scoreR0, returnalignedSeq1, returnalignedSeq2, starti, startj, endi, endj = cleaner.SWAlignment(
tmpread, rightSeg[0], parameterRobot)
scoreR1, returnalignedSeq1, returnalignedSeq2, starti, startj, endi, endj = cleaner.SWAlignment(
tmpread, rightSeg[1], parameterRobot)
skipCounting = False
# Filering reads
if startj > threshold or endj < len(leftSeg[0]) - threshold:
skipCounting = True
# MAP decision rule
if not skipCounting:
MAPScoreL = [0, 0]
for i in range(2):
#tempScore, returnalignedSeq1, returnalignedSeq2 , starti, startj , endi, endj, numPath= cleaner.SWAlignmentFixRefMP(leftSeg[i],tmpread, parameterRobot)
#lengthSeed = len(leftSeg[i])
#countConfirm, countDel, countIns, countSub = cleaner.countEdits(returnalignedSeq2, returnalignedSeq1) # care for the 1,2 and its fcn
#confirmNoDelete,confirmNoInsert = countConfirm , lengthSeed - countIns
#MAPScoreL[i] = math.log(1-perr)*confirmNoDelete + math.log(1-perr)*confirmNoInsert + math.log(perr)*countDel + math.log(perr*perr/3)*countSub + math.log(perr/3)*countIns + math.log(numPath)
tempScore, returnalignedSeq1, returnalignedSeq2, starti, startj, endi, endj, numPath = cleaner.SWAlignmentFixRefMPQS(
leftSeg[i], tmpread, parameterRobot)
MAPScoreL[i] = tempScore
MAPScoreR = [0, 0]
for i in range(2):
#tempScore, returnalignedSeq1, returnalignedSeq2 , starti, startj , endi, endj, numPath= cleaner.SWAlignmentFixRefMP(rightSeg[i],tmpread, parameterRobot)
#lengthSeed = len(leftSeg[i])
#countConfirm, countDel, countIns, countSub = cleaner.countEdits(returnalignedSeq2, returnalignedSeq1) # care for the 1,2 and its fcn
#confirmNoDelete,confirmNoInsert = countConfirm , lengthSeed - countIns
#MAPScoreR[i] = math.log(1-perr)*confirmNoDelete + math.log(1-perr)*confirmNoInsert + math.log(perr)*countDel + math.log(perr*perr/3)*countSub + math.log(perr/3)*countIns + math.log(numPath)
tempScore, returnalignedSeq1, returnalignedSeq2, starti, startj, endi, endj, numPath = cleaner.SWAlignmentFixRefMPQS(
rightSeg[i], tmpread, parameterRobot)
MAPScoreR[i] = tempScore
H0Score = max(MAPScoreL[0] + MAPScoreR[0],
MAPScoreL[1] + MAPScoreR[1])
H1Score = max(MAPScoreL[1] + MAPScoreR[0],
MAPScoreL[0] + MAPScoreR[1])
H0 = H0 + H0Score
H1 = H1 + H1Score
if H0 > H1:
middleMap = [0, 1]
else:
middleMap = [1, 0]
print "Middle : H0, H1", H0, H1
return middleMap, H0, H1
def decideMiddlePiece(commonList, leftSeg, rightSeg, parameterRobot):
middleMap = []
threshold = 10
perr = parameterRobot.p
H0 , H1 = 0 ,0
for i in range(len(commonList)):
read = commonList[i].longread
score, returnalignedSeq1, returnalignedSeq2 , starti, startj , endi, endj = cleaner.SWAlignment(read, leftSeg[0], parameterRobot)
readRev = cleaner.reverseStrand(read)
scoreRev, returnalignedSeq1Rev, returnalignedSeq2Rev , startiRev, startjRev , endiRev, endjRev = cleaner.SWAlignment(readRev, leftSeg[0], parameterRobot)
if score > scoreRev :
forwardCk = True
scoreL0 = score
else:
forwardCk = False
scoreL0 = scoreRev
if forwardCk :
tmpread = read
else:
tmpread = readRev
scoreL1, returnalignedSeq1, returnalignedSeq2 , starti, startj , endi, endj = cleaner.SWAlignment(tmpread, leftSeg[1], parameterRobot)
scoreR0 , returnalignedSeq1, returnalignedSeq2 , starti, startj , endi, endj = cleaner.SWAlignment(tmpread, rightSeg[0], parameterRobot)
scoreR1 , returnalignedSeq1, returnalignedSeq2 , starti, startj , endi, endj = cleaner.SWAlignment(tmpread, rightSeg[1], parameterRobot)
skipCounting = False
# Filering reads
if startj > threshold or endj < len(leftSeg[0]) - threshold:
skipCounting = True
# MAP decision rule
if not skipCounting:
MAPScoreL= [0,0]
for i in range(2):
#tempScore, returnalignedSeq1, returnalignedSeq2 , starti, startj , endi, endj, numPath= cleaner.SWAlignmentFixRefMP(leftSeg[i],tmpread, parameterRobot)
#lengthSeed = len(leftSeg[i])
#countConfirm, countDel, countIns, countSub = cleaner.countEdits(returnalignedSeq2, returnalignedSeq1) # care for the 1,2 and its fcn
#confirmNoDelete,confirmNoInsert = countConfirm , lengthSeed - countIns
#MAPScoreL[i] = math.log(1-perr)*confirmNoDelete + math.log(1-perr)*confirmNoInsert + math.log(perr)*countDel + math.log(perr*perr/3)*countSub + math.log(perr/3)*countIns + math.log(numPath)
tempScore, returnalignedSeq1, returnalignedSeq2 , starti, startj , endi, endj, numPath= cleaner.SWAlignmentFixRefMPQS(leftSeg[i],tmpread, parameterRobot)
MAPScoreL[i] =tempScore
MAPScoreR= [0,0]
for i in range(2):
#tempScore, returnalignedSeq1, returnalignedSeq2 , starti, startj , endi, endj, numPath= cleaner.SWAlignmentFixRefMP(rightSeg[i],tmpread, parameterRobot)
#lengthSeed = len(leftSeg[i])
#countConfirm, countDel, countIns, countSub = cleaner.countEdits(returnalignedSeq2, returnalignedSeq1) # care for the 1,2 and its fcn
#confirmNoDelete,confirmNoInsert = countConfirm , lengthSeed - countIns
#MAPScoreR[i] = math.log(1-perr)*confirmNoDelete + math.log(1-perr)*confirmNoInsert + math.log(perr)*countDel + math.log(perr*perr/3)*countSub + math.log(perr/3)*countIns + math.log(numPath)
tempScore, returnalignedSeq1, returnalignedSeq2 , starti, startj , endi, endj, numPath= cleaner.SWAlignmentFixRefMPQS(rightSeg[i],tmpread, parameterRobot)
MAPScoreR[i] = tempScore
H0Score = max( MAPScoreL[0] + MAPScoreR[0] , MAPScoreL[1] + MAPScoreR[1])
H1Score = max( MAPScoreL[1] + MAPScoreR[0], MAPScoreL[0] + MAPScoreR[1])
H0 = H0 + H0Score
H1 = H1 + H1Score
if H0 > H1:
middleMap = [0,1]
else:
middleMap = [1,0]
print "Middle : H0, H1", H0, H1
return middleMap , H0, H1
def findVoteScore(in1List, leftSeg, parameterRobot):
H0Score, H1Score = 0,0
threshold = 10
perr = parameterRobot.p
print " len(in1List) ", len(in1List)
for i in range(len(in1List)):
# Determine F/B
read = in1List[i].longread
score, returnalignedSeq1, returnalignedSeq2 , starti, startj , endi, endj = cleaner.SWAlignment(read, leftSeg[0], parameterRobot)
reverseread = cleaner.reverseStrand(read)
scoreRev, returnalignedSeq1Rev, returnalignedSeq2Rev , startiRev, startjRev , endiRev, endjRev = cleaner.SWAlignment(reverseread, leftSeg[0], parameterRobot)
if score > scoreRev:
forwardCk = True
scoreLeft0 = score
else:
forwardCk =False
scoreLeft0 = scoreRev
if forwardCk :
score, returnalignedSeq1, returnalignedSeq2 , starti, startj , endi, endj = cleaner.SWAlignment(read, leftSeg[1], parameterRobot)
scoreLeft1 = score
else:
score, returnalignedSeq1, returnalignedSeq2 , starti, startj , endi, endj = cleaner.SWAlignment(reverseread, leftSeg[1], parameterRobot)
scoreLeft1 = score
print "scoreLeft0, scoreLeft1 " , scoreLeft0, scoreLeft1
skipCounting = False
MAPScore= [0,0]
# Filering reads
if startj > threshold or endj < len(leftSeg[0]) - threshold:
skipCounting = True
# MAP decision rule
if forwardCk :
for i in range(2):
#tempScore, returnalignedSeq1, returnalignedSeq2 , starti, startj , endi, endj, numPath= cleaner.SWAlignmentFixRefMP(leftSeg[i],read, parameterRobot)
#lengthSeed = len(leftSeg[i])
#countConfirm, countDel, countIns, countSub = cleaner.countEdits(returnalignedSeq2, returnalignedSeq1) # care for the 1,2 and its fcn
#confirmNoDelete,confirmNoInsert = countConfirm , lengthSeed - countIns
#MAPScore[i] = math.log(1-perr)*confirmNoDelete + math.log(1-perr)*confirmNoInsert + math.log(perr)*countDel + math.log(perr*perr/3)*countSub + math.log(perr/3)*countIns + math.log(numPath)
tempScore, returnalignedSeq1, returnalignedSeq2 , starti, startj , endi, endj, numPath= cleaner.SWAlignmentFixRefMPQS(leftSeg[i],read, parameterRobot)
MAPScore[i] =tempScore
else:
for i in range(2):
#tempScore, returnalignedSeq1, returnalignedSeq2 , starti, startj , endi, endj, numPath= cleaner.SWAlignmentFixRefMP(leftSeg[i],reverseread, parameterRobot)
#lengthSeed = len(leftSeg[i])
#countConfirm, countDel, countIns, countSub = cleaner.countEdits(returnalignedSeq2, returnalignedSeq1) # care for the 1,2 and its fcn
#confirmNoDelete,confirmNoInsert = countConfirm , lengthSeed - countIns
#MAPScore[i] = math.log(1-perr)*confirmNoDelete + math.log(1-perr)*confirmNoInsert + math.log(perr)*countDel + math.log(perr*perr/3)*countSub + math.log(perr/3)*countIns + math.log(numPath)
tempScore, returnalignedSeq1, returnalignedSeq2 , starti, startj , endi, endj, numPath= cleaner.SWAlignmentFixRefMPQS(leftSeg[i],reverseread, parameterRobot)
MAPScore[i] = tempScore
if not skipCounting:
H0Score += MAPScore[0]
H1Score += MAPScore[1]
return H0Score , H1Score
def linkSegments(inSeg, middleSeg, outSeg, parameterRobot):
print "\nNew Contig"
middleRead = middleSeg.longread
revmiddleRead = cleaner.reverseStrand(middleRead)
extendedRead = []
score, scoreRev = -1, -1
inIndex = 0
startjTmp = -1
while ((max(score, scoreRev)< 30 or startjTmp> 10) and inIndex < len(inSeg)):
inRead = inSeg[inIndex].longread
revinRead = cleaner.reverseStrand(inRead)
score, returnalignedSeq1, returnalignedSeq2 , starti, startj , endi, endj = cleaner.SWAlignment(inRead, middleRead, parameterRobot)
scoreRev, returnalignedSeq1Rev, returnalignedSeq2Rev , startiRev, startjRev , endiRev, endjRev = cleaner.SWAlignment(revinRead, middleRead, parameterRobot)
if score > scoreRev:
startjTmp = startj
print "starti, startj , endi, endj", starti, startj , endi, endj
else:
startjTmp = startjRev
print "startiRev, startjRev , endiRev, endjRev", startiRev, startjRev , endiRev, endjRev
print "startjTmp", startjTmp
print "score, scoreRev,inIndex " ,score, scoreRev ,inIndex
inIndex += 1
if score > scoreRev :
for eachbase in inRead[0:starti]:
extendedRead.append(eachbase)
print "len(extendedRead)", len(extendedRead)
for eachbase in middleRead:
extendedRead.append(eachbase)
print "len(extendedRead)", len(extendedRead)
else:
for eachbase in revinRead[0:startiRev]:
extendedRead.append(eachbase)
print "len(extendedRead)", len(extendedRead)
for eachbase in middleRead:
extendedRead.append(eachbase)
print "len(extendedRead)", len(extendedRead)
score, scoreRev = -1, -1
# hack
outIndex = 0
endiTmp = 1000
outRead, revoutRead = [] , []
while ((max(score, scoreRev)< 30 or endiTmp < len(middleRead)-10)and outIndex < len(outSeg)):
outRead = outSeg[outIndex].longread
revoutRead = cleaner.reverseStrand(outRead)
score, returnalignedSeq1, returnalignedSeq2 , starti, startj , endi, endj = cleaner.SWAlignment(middleRead,outRead , parameterRobot)
scoreRev, returnalignedSeq1Rev, returnalignedSeq2Rev , startiRev, startjRev , endiRev, endjRev = cleaner.SWAlignment(middleRead, revoutRead, parameterRobot)
if score > scoreRev :
endiTmp = endi
print "starti, startj , endi, endj", starti, startj , endi, endj
else:
endiTmp = endiRev
print " startiRev, startjRev , endiRev, endjRev ", startiRev, startjRev , endiRev, endjRev
print "score, scoreRev, outIndex" ,score, scoreRev, outIndex
outIndex += 1
correctionTerm = len(middleRead) - endiTmp
if score > scoreRev :
for eachbase in outRead[endj+correctionTerm:len(outRead)]:
extendedRead.append(eachbase)
print "len(extendedRead) forw", len(extendedRead)
else:
for eachbase in revoutRead[endjRev+correctionTerm:len(revoutRead)]:
extendedRead.append(eachbase)
print "len(extendedRead) rev", len(extendedRead)
return extendedRead