def createIndelRobot(folderName):
indelRobot = common.parameterRobot()
indelRobot.defaultFolder = folderName
# indelRobot.setReadStat( Nshort= parameterRobot.N, Nlong= parameterRobot.N, Lshort= parameterRobot.L, Llong= parameterRobot.L, p= parameterRobot.p , longOnly = True)
# indelRobot.setGenomeStat(G = parameterRobot.G, lrep=500, lsnp=200, lint=50 )
indelRobot.setThresholdPara(liid=30, thresForRandom=0.5, thresForins=0.4, thresFordel=0.4, insMin=4, delMin=4, thresholdForSupport=0.15, subthreshold=9, editsub=-10, editins=-1, editdel=-1, editmatch=1, lookRange=15)
indelRobot.tunePara()
indelRobot.snprate = 0.01
return indelRobot
def arrangeSeqBasedOnRefEasy(motherGenome, reconstructedGenome,
parameterRobot):
runningI = 0
runningJ = 0
G = len(motherGenome)
W = 50
totalScore = 0
parameterRobot = common.parameterRobot()
counter = 0
scoreList = []
while runningI < len(motherGenome) - 1 and counter < parameterRobot.G * 1.1:
score, returnalignedSeq1, returnalignedSeq2, starti, startj, endi, endj = cleaner.SWAlignment(
motherGenome[runningI:runningI + W],
reconstructedGenome[runningJ:runningJ + W], parameterRobot)
scoreList.append([score, starti, startj, endi, endj, runningI])
runningI = runningI + W
counter = counter + 1
scoreList = sorted(scoreList)
testCases = []
for i in range(3):
score, starti, startj, endi, endj, runningI = scoreList[-(i + 1)]
newGenome = np.zeros(G, dtype=np.int32)
myindex = runningI + starti - startj - runningJ
if myindex > 0:
newGenome[0:G - myindex] = motherGenome[myindex:G]
newGenome[G - myindex:G] = motherGenome[0:myindex]
else:
myposindex = G + myindex
newGenome[0:G - myposindex] = motherGenome[myposindex:G]
newGenome[G - myposindex:G] = motherGenome[0:myposindex]
testCases.append(newGenome)
return testCases
def findMismatchNumber(motherGenome, reconstructedGenome):
## Fill in the algorithm here
# 1) Sliding window matching and count
# 2) Make sure to not include the loop around thing.
W = 50
runningI = 0
runningJ = 0
G = len(motherGenome)
totalScore = 0
parameterRobot = common.parameterRobot()
counter = 0
tmprunningI = -1
print "----------------------"
overallSum = 0
while runningI < len(motherGenome) - 1 and tmprunningI != runningI:
score, returnalignedSeq1, returnalignedSeq2, starti, startj, endi, endj = cleaner.SWAlignmentFixRef(
motherGenome[runningI:runningI + W],
reconstructedGenome[runningJ:runningJ + W], parameterRobot)
#print "Mother: "
#cluster.printSeq(returnalignedSeq1)
#print "Reconstructed: "
#cluster.printSeq(returnalignedSeq2)
#print "----------"
tmprunningI = runningI
runningI = endi + runningI
runningJ = endj + runningJ
counter = counter + 1
totalScore = totalScore + score
#print "score", score
overallSum += score
print "----------------------"
numberOfMismatch = max((G - totalScore) / 2, 0)
print overallSum / counter
return numberOfMismatch
def findMismatchNumber(motherGenome, reconstructedGenome):
## Fill in the algorithm here
# 1) Sliding window matching and count
# 2) Make sure to not include the loop around thing.
W= 50
runningI = 0
runningJ = 0
G = len(motherGenome)
totalScore = 0
parameterRobot = common.parameterRobot()
counter = 0
tmprunningI = -1
print "----------------------"
overallSum = 0
while runningI < len(motherGenome) -1 and tmprunningI != runningI:
score, returnalignedSeq1, returnalignedSeq2 , starti, startj , endi, endj = cleaner.SWAlignmentFixRef(motherGenome[runningI:runningI+W] , reconstructedGenome[runningJ:runningJ+W], parameterRobot)
#print "Mother: "
#cluster.printSeq(returnalignedSeq1)
#print "Reconstructed: "
#cluster.printSeq(returnalignedSeq2)
#print "----------"
tmprunningI = runningI
runningI = endi + runningI
runningJ = endj + runningJ
counter = counter +1
totalScore = totalScore + score
#print "score", score
overallSum += score
print "----------------------"
numberOfMismatch = max((G - totalScore)/2, 0 )
print overallSum / counter
return numberOfMismatch
def arrangeSeqBasedOnRefEasy(motherGenome, reconstructedGenome, parameterRobot):
runningI = 0
runningJ = 0
G = len(motherGenome)
W = 50
totalScore = 0
parameterRobot = common.parameterRobot()
counter = 0
scoreList = []
while runningI < len(motherGenome) -1 and counter < parameterRobot.G*1.1:
score, returnalignedSeq1, returnalignedSeq2 , starti, startj , endi, endj = cleaner.SWAlignment(motherGenome[runningI:runningI+W] , reconstructedGenome[runningJ:runningJ+W], parameterRobot)
scoreList.append([score,starti, startj , endi, endj,runningI ])
runningI = runningI + W
counter = counter +1
scoreList = sorted(scoreList)
testCases = []
for i in range(3):
score, starti, startj , endi, endj,runningI = scoreList[-(i+1)]
newGenome = np.zeros(G, dtype = np.int32)
myindex = runningI +starti - startj-runningJ
if myindex > 0 :
newGenome[0:G-myindex] = motherGenome[myindex:G]
newGenome[G-myindex:G] = motherGenome[0:myindex]
else:
myposindex = G+ myindex
newGenome[0:G-myposindex] = motherGenome[myposindex:G]
newGenome[G-myposindex:G] = motherGenome[0:myposindex]
testCases.append(newGenome)
return testCases
def indelMSABridging(f2, currentNode,noisyReads,p, snpRate,flankinglen, parameterRobot):
# Need to loop over all the xnodes
canResolve, kmerPairsList = False, []
# Using contig Creator
indelRobot = common.parameterRobot()
indelRobot.defaultFolder = parameterRobot.defaultFolder
indelRobot.setReadStat( Nshort= parameterRobot.N, Nlong= parameterRobot.N, Lshort= parameterRobot.L, Llong= parameterRobot.L, p= parameterRobot.p , longOnly = True)
indelRobot.setGenomeStat(G = parameterRobot.G, lrep=500, lsnp=200, lint=50 )
indelRobot.setThresholdPara(liid = 30, thresForRandom= 0.5,thresForins =0.4, thresFordel=0.4, insMin=4, delMin=4,thresholdForSupport= 0.15, subthreshold= 9, editsub= -10, editins= -1, editdel= -1, editmatch = 1, lookRange =15)
#indelRobot.tunePara()
indelRobot.snprate = snpRate
# toProcessList : in1IndexList, in2IndexList, out1IndexList, out2IndexList, commonIndexList
# shortToLongMap : indexlong indexshort jstart jend istart iend
shortToLongMap,toProcessList = [], []
toProcessList = formToProcessList(f2, noisyReads, currentNode, indelRobot, flankinglen)
if len(toProcessList[4] ) == 0 :
return False, []
shortToLongMap = formRelatedMap(f2, noisyReads, currentNode, indelRobot, toProcessList)
cleaner.cleaning([noisyReads,noisyReads] ,shortToLongMap, toProcessList,indelRobot, "init")
in1List, in2List, out1List, out2List, commonList, longReadToUse = cleaner.cleaning([noisyReads, noisyReads],shortToLongMap, toProcessList,indelRobot, "vote")
extendResult = extender.readExtender(in1List, in2List, out1List, out2List, commonList,indelRobot,longReadToUse, True)
if extendResult == 0 :
canResolve = True
edgeWt = currentNode.listOfPrevNodes[0][1]
inKmerIndex = currentNode.listOfPrevNodes[0][0].nodeIndexList[-(edgeWt +1)]
edgeWt = currentNode.listOfNextNodes[0][1]
outKmerIndex = currentNode.listOfNextNodes[0][0].nodeIndexList[edgeWt ]
kmerPairsList.append([inKmerIndex,outKmerIndex ])
edgeWt = currentNode.listOfPrevNodes[1][1]
inKmerIndex = currentNode.listOfPrevNodes[1][0].nodeIndexList[-(edgeWt +1)]
edgeWt = currentNode.listOfNextNodes[1][1]
outKmerIndex = currentNode.listOfNextNodes[1][0].nodeIndexList[edgeWt ]
kmerPairsList.append([inKmerIndex,outKmerIndex ])
elif extendResult == 1 :
canResolve = True
edgeWt = currentNode.listOfPrevNodes[0][1]
inKmerIndex = currentNode.listOfPrevNodes[0][0].nodeIndexList[-(edgeWt +1)]
edgeWt = currentNode.listOfNextNodes[1][1]
outKmerIndex = currentNode.listOfNextNodes[1][0].nodeIndexList[edgeWt ]
kmerPairsList.append([inKmerIndex,outKmerIndex ])
edgeWt = currentNode.listOfPrevNodes[1][1]
inKmerIndex = currentNode.listOfPrevNodes[1][0].nodeIndexList[-(edgeWt +1)]
edgeWt = currentNode.listOfNextNodes[0][1]
outKmerIndex = currentNode.listOfNextNodes[0][0].nodeIndexList[edgeWt ]
kmerPairsList.append([inKmerIndex,outKmerIndex ])
elif extendResult == -1:
canResolve = False
kmerPairsList = []
print kmerPairsList
return canResolve, kmerPairsList
def indelMSABridging(f2, currentNode, noisyReads, p, snpRate, flankinglen,
parameterRobot):
# Need to loop over all the xnodes
canResolve, kmerPairsList = False, []
# Using contig Creator
indelRobot = common.parameterRobot()
indelRobot.defaultFolder = parameterRobot.defaultFolder
indelRobot.setReadStat(Nshort=parameterRobot.N,
Nlong=parameterRobot.N,
Lshort=parameterRobot.L,
Llong=parameterRobot.L,
p=parameterRobot.p,
longOnly=True)
indelRobot.setGenomeStat(G=parameterRobot.G, lrep=500, lsnp=200, lint=50)
indelRobot.setThresholdPara(liid=30,
thresForRandom=0.5,
thresForins=0.4,
thresFordel=0.4,
insMin=4,
delMin=4,
thresholdForSupport=0.15,
subthreshold=9,
editsub=-10,
editins=-1,
editdel=-1,
editmatch=1,
lookRange=15)
#indelRobot.tunePara()
indelRobot.snprate = snpRate
# toProcessList : in1IndexList, in2IndexList, out1IndexList, out2IndexList, commonIndexList
# shortToLongMap : indexlong indexshort jstart jend istart iend
shortToLongMap, toProcessList = [], []
toProcessList = formToProcessList(f2, noisyReads, currentNode, indelRobot,
flankinglen)
if len(toProcessList[4]) == 0:
return False, []
shortToLongMap = formRelatedMap(f2, noisyReads, currentNode, indelRobot,
toProcessList)
cleaner.cleaning([noisyReads, noisyReads], shortToLongMap, toProcessList,
indelRobot, "init")
in1List, in2List, out1List, out2List, commonList, longReadToUse = cleaner.cleaning(
[noisyReads, noisyReads], shortToLongMap, toProcessList, indelRobot,
"vote")
extendResult = extender.readExtender(in1List, in2List, out1List, out2List,
commonList, indelRobot, longReadToUse,
True)
if extendResult == 0:
canResolve = True
edgeWt = currentNode.listOfPrevNodes[0][1]
inKmerIndex = currentNode.listOfPrevNodes[0][0].nodeIndexList[-(
edgeWt + 1)]
edgeWt = currentNode.listOfNextNodes[0][1]
outKmerIndex = currentNode.listOfNextNodes[0][0].nodeIndexList[edgeWt]
kmerPairsList.append([inKmerIndex, outKmerIndex])
edgeWt = currentNode.listOfPrevNodes[1][1]
inKmerIndex = currentNode.listOfPrevNodes[1][0].nodeIndexList[-(
edgeWt + 1)]
edgeWt = currentNode.listOfNextNodes[1][1]
outKmerIndex = currentNode.listOfNextNodes[1][0].nodeIndexList[edgeWt]
kmerPairsList.append([inKmerIndex, outKmerIndex])
elif extendResult == 1:
canResolve = True
edgeWt = currentNode.listOfPrevNodes[0][1]
inKmerIndex = currentNode.listOfPrevNodes[0][0].nodeIndexList[-(
edgeWt + 1)]
edgeWt = currentNode.listOfNextNodes[1][1]
outKmerIndex = currentNode.listOfNextNodes[1][0].nodeIndexList[edgeWt]
kmerPairsList.append([inKmerIndex, outKmerIndex])
edgeWt = currentNode.listOfPrevNodes[1][1]
inKmerIndex = currentNode.listOfPrevNodes[1][0].nodeIndexList[-(
edgeWt + 1)]
edgeWt = currentNode.listOfNextNodes[0][1]
outKmerIndex = currentNode.listOfNextNodes[0][0].nodeIndexList[edgeWt]
kmerPairsList.append([inKmerIndex, outKmerIndex])
elif extendResult == -1:
canResolve = False
kmerPairsList = []
print kmerPairsList
return canResolve, kmerPairsList