def transformReads(G, N, L, folderName, myIndex):
motherGen, reads, noisyReads = logging.rawDataLoad(folderName + "UnitTest",
G, N, L, "dn")
myRead = reads[myIndex]
fout = open(folderName + "bridgingReads_" + str(myIndex) + ".fasta", 'w')
fout.write(">Seg1\n")
for i in range(len(myRead)):
if myRead[i] == 1:
fout.write('A')
elif myRead[i] == 2:
fout.write('C')
elif myRead[i] == 3:
fout.write('G')
elif myRead[i] == 4:
fout.write('T')
#else:
# fout.write('A')
if np.mod(i, 70) == 69:
fout.write("\n")
fout.close()
print "len(motherGen) , len(reads), len(noisyReads) : ", len(
motherGen), len(reads), len(noisyReads)
def branchClearingUnitTest(G, N, L,K, liid, threshold, foldername = "",branchingDepth= 20):
dummyParameters = logging.parameterObj()
dummyParameters.defaultFolder = foldername
dummyParameters.brachingDepth = branchingDepth
dummyParameters.K, dummyParameters.liid, dummyParameters.threshold = K, liid, threshold
dummyParameters.G, dummyParameters.N, dummyParameters.L = G, N, L
motherGen, reads, noisyReads = logging.rawDataLoad(foldername+"UnitTest",G,N,L, "dn")
returnfmapping= logging.fmappingLoad(foldername+'clusteredGroup.csv')
G1,startList, fmapping = graphForm.getSeqGraph(returnfmapping,noisyReads, dummyParameters)
#checkCondensingBasic(G1, startList, "simple")
returnfmapping, G1= branchClear.clearResidual(returnfmapping,G1,dummyParameters)
G2 = G1
graphForm.debugSeqGraph(G2)
#G2 = G1
checkCondensingBasic2(G2, [G2[0]], "simple")
def branchClearingUnitTest(G,
N,
L,
K,
liid,
threshold,
foldername="",
branchingDepth=20):
dummyParameters = logging.parameterObj()
dummyParameters.defaultFolder = foldername
dummyParameters.brachingDepth = branchingDepth
dummyParameters.K, dummyParameters.liid, dummyParameters.threshold = K, liid, threshold
dummyParameters.G, dummyParameters.N, dummyParameters.L = G, N, L
motherGen, reads, noisyReads = logging.rawDataLoad(foldername + "UnitTest",
G, N, L, "dn")
returnfmapping = logging.fmappingLoad(foldername + 'clusteredGroup.csv')
G1, startList, fmapping = graphForm.getSeqGraph(returnfmapping, noisyReads,
dummyParameters)
#checkCondensingBasic(G1, startList, "simple")
returnfmapping, G1 = branchClear.clearResidual(returnfmapping, G1,
dummyParameters)
G2 = G1
graphForm.debugSeqGraph(G2)
#G2 = G1
checkCondensingBasic2(G2, [G2[0]], "simple")
def ECUnitTest(Nin,G, Lin, foldername = "", bridgingDepth = 20, msaWidth = 20 ):
dummyParameters = logging.parameterObj()
dummyParameters.bridgingDepth = bridgingDepth
dummyParameters.msaWidth = msaWidth
dummyParameters.N, dummyParameters.G, dummyParameters.L, dummyParameters.p = Nin, G,Lin, 0.015
dummyParameters.indel = True
dummyParameters.defaultFolder = foldername
dummyParameters.threshold = 5
dummyParameters.liid = 48
snpRate, typeOfGen, detail = 0.001 ,'m', "500-200-50"
G,N,L = dummyParameters.G, dummyParameters.N, dummyParameters.L,
motherGen, reads, noisyReads = logging.rawDataLoad(foldername+"UnitTest",G,N,L, "dn")
f2= logging.fmapfusedLoad(foldername+'clusteredGroup2.csv')
G2 = logging.loadGraph(foldername+'basicMapping.csv', foldername+'seqMapping.txt', 'simple')
checkCondensingBasic(G2, [G2[0]], "simple")
G3 = bridgeResolve.resolveRepeats(f2,G2,dummyParameters)
checkCondensingBasic(G3, [G3[0]], "MB")
G4 = alignmentBridge.MSAresolve(f2, G3, noisyReads, snpRate,dummyParameters)
#G4 = G3
#checkCondensingBasic(G4, [G4[0]], "MB")
recovSeq = eulerCycle.findEC(G4)
recovGen = readAns.reportRecovSeq(recovSeq, f2, noisyReads,dummyParameters)
numMistakes, success = compare.subAlignCompare(recovGen, motherGen,dummyParameters)
return numMistakes, success
def transformReads(G,N,L,folderName,myIndex):
motherGen, reads, noisyReads = logging.rawDataLoad(folderName+"UnitTest",G,N,L, "dn")
myRead = reads[myIndex]
fout= open(folderName + "bridgingReads_"+str(myIndex)+".fasta", 'w')
fout.write(">Seg1\n")
for i in range(len(myRead)):
if myRead[i] ==1 :
fout.write('A')
elif myRead[i] ==2 :
fout.write('C')
elif myRead[i] ==3 :
fout.write('G')
elif myRead[i] ==4 :
fout.write('T')
#else:
# fout.write('A')
if np.mod(i,70) == 69 :
fout.write("\n")
fout.close()
print "len(motherGen) , len(reads), len(noisyReads) : ", len(motherGen) , len(reads), len(noisyReads)
def runAssembler(snpRate, typeOfGen, detail, parameterRobot):
#N, G, L, p,K,snpRate, typeOfGen, detail = 100, 100, 10000, 0.01,30,0.001, 'r', "1000"
motherGen, reads, noisyReads = dataGen.generateData(
typeOfGen, detail, parameterRobot)
motherGen, reads, noisyReads = logging.rawDataLoad(
parameterRobot.defaultFolder + "UnitTest", parameterRobot.G,
parameterRobot.N, parameterRobot.L, "dn")
f1 = cluster.groupIndelNoisyKmers(noisyReads, parameterRobot)
G1, startList, f1 = graphForm.getSeqGraph(f1, noisyReads, parameterRobot)
f2, G2 = branchClear.clearResidual(f1, G1, parameterRobot)
G3 = bridgeResolve.resolveRepeats(f2, G2, parameterRobot)
G4 = alignmentBridge.MSAresolve(f2, G3, noisyReads, snpRate,
parameterRobot)
#G4 = G3
recovSeq = eulerCycle.findEC(G4)
recovGen = readAns.reportRecovSeq(recovSeq, f2, noisyReads, parameterRobot)
numMistakes, success = compare.subAlignCompare(recovGen, motherGen,
parameterRobot)
#numMistakes, success = 0 , 0
return numMistakes, success
# Target of the new code : Fast Assemble and assembly in optimal amount of information
#t0 = time.time()
#N, G, L, p,snpRate, typeOfGen, detail = 1000, 10000,200, 0.015, 0.001 ,'m', "500-300-50"
#runAssembler(N, G, L, p,snpRate, typeOfGen, detail)
#print "Time (sec) :", time.time() - t0
def runAssembler(snpRate, typeOfGen, detail,parameterRobot):
#N, G, L, p,K,snpRate, typeOfGen, detail = 100, 100, 10000, 0.01,30,0.001, 'r', "1000"
motherGen, reads, noisyReads = dataGen.generateData(typeOfGen, detail,parameterRobot)
motherGen, reads, noisyReads = logging.rawDataLoad(parameterRobot.defaultFolder+"UnitTest",parameterRobot.G,parameterRobot.N,parameterRobot.L, "dn")
f1 = cluster.groupIndelNoisyKmers(noisyReads,parameterRobot)
G1,startList, f1 = graphForm.getSeqGraph(f1,noisyReads, parameterRobot)
f2, G2 = branchClear.clearResidual(f1, G1,parameterRobot)
G3 = bridgeResolve.resolveRepeats(f2, G2,parameterRobot)
G4 = alignmentBridge.MSAresolve(f2, G3, noisyReads, snpRate,parameterRobot )
#G4 = G3
recovSeq = eulerCycle.findEC(G4)
recovGen = readAns.reportRecovSeq(recovSeq, f2, noisyReads,parameterRobot)
numMistakes, success = compare.subAlignCompare(recovGen, motherGen,parameterRobot)
#numMistakes, success = 0 , 0
return numMistakes, success
# Target of the new code : Fast Assemble and assembly in optimal amount of information
#t0 = time.time()
#N, G, L, p,snpRate, typeOfGen, detail = 1000, 10000,200, 0.015, 0.001 ,'m', "500-300-50"
#runAssembler(N, G, L, p,snpRate, typeOfGen, detail)
#print "Time (sec) :", time.time() - t0
def graphFormUnitTest2(N, G, L, folderName = ""):
dummyParameters = logging.parameterObj()
dummyParameters.defaultFolder = folderName
returnfmapping= logging.fmappingLoad(dummyParameters.defaultFolder+'clusteredGroup.csv')
dummyParameters.N, dummyParameters.G, dummyParameters.L, dummyParameters.p, typeOfGen, detail = N, G, L, 0.015, 'm', "500-200-50"
motherGen, reads, noisyReads = logging.rawDataLoad(dummyParameters.defaultFolder+"UnitTest",dummyParameters.G,dummyParameters.N,dummyParameters.L, "dn")
G1,startList,fmapping = graphForm.getSeqGraph(returnfmapping,noisyReads, dummyParameters )
def clusteringTestingOfK(G, N, L, folderName, K, threshold, liid):
motherGen, reads, noisyReads = logging.rawDataLoad(folderName + "UnitTest",G,N,L,'a')
dummyParameters = logging.parameterObj()
dummyParameters.brachingDepth = 20
dummyParameters.clusterRounds , dummyParameters.fingerPrint, dummyParameters.clusterRatio = 2, 6, 2
dummyParameters.N, dummyParameters.L, dummyParameters.K, dummyParameters.G, dummyParameters.threshold, dummyParameters.liid = N, L, K, G, threshold, liid
cluster.groupNoisyKmers(noisyReads,dummyParameters, 'fast' )
def clusteringTestingOfK(G, N, L, folderName, K, threshold, liid):
motherGen, reads, noisyReads = logging.rawDataLoad(folderName + "UnitTest",
G, N, L, 'a')
dummyParameters = logging.parameterObj()
dummyParameters.brachingDepth = 20
dummyParameters.clusterRounds, dummyParameters.fingerPrint, dummyParameters.clusterRatio = 2, 6, 2
dummyParameters.N, dummyParameters.L, dummyParameters.K, dummyParameters.G, dummyParameters.threshold, dummyParameters.liid = N, L, K, G, threshold, liid
cluster.groupNoisyKmers(noisyReads, dummyParameters, 'fast')
def graphFormUnitTest2(N, G, L, folderName=""):
dummyParameters = logging.parameterObj()
dummyParameters.defaultFolder = folderName
returnfmapping = logging.fmappingLoad(dummyParameters.defaultFolder +
'clusteredGroup.csv')
dummyParameters.N, dummyParameters.G, dummyParameters.L, dummyParameters.p, typeOfGen, detail = N, G, L, 0.015, 'm', "500-200-50"
motherGen, reads, noisyReads = logging.rawDataLoad(
dummyParameters.defaultFolder + "UnitTest", dummyParameters.G,
dummyParameters.N, dummyParameters.L, "dn")
G1, startList, fmapping = graphForm.getSeqGraph(returnfmapping, noisyReads,
dummyParameters)
def graphFormUnitTest(N, G, L, folderName =""):
print "dataGenUnitTest"
dummyParameters = logging.parameterObj()
dummyParameters.defaultFolder = folderName
dummyParameters.liid = 40
dummyParameters.K = 40
dummyParameters.threshold = 6
dummyParameters.p = 0.015
dummyParameters.indel = True
dummyParameters.N, dummyParameters.G, dummyParameters.L, dummyParameters.p, typeOfGen, detail = N, G, L, 0.015, 'm', "500-200-50"
# motherGen, reads, noisyReads = dataGen.generateData( typeOfGen,detail,dummyParameters)
motherGen, reads, noisyReads = logging.rawDataLoad(dummyParameters.defaultFolder+"UnitTest",dummyParameters.G,dummyParameters.N,dummyParameters.L, "dn")
returnfmapping= cluster.groupIndelNoisyKmers(noisyReads, dummyParameters, "fast")
def clusterUnitTest2():
# Unit Test 1 : Generate 30 reads with 10 copies with noise , length being 20
motherGen, reads, noisyReads = logging.rawDataLoad("clusterReadsUnitTest",
10, 300, 40, "dn")
parameterRobot = logging.parameterObj()
parameterRobot.N = 300
parameterRobot.L = 40
parameterRobot.G = 10000
parameterRobot.liid = 30
parameterRobot.K = 30
parameterRobot.threshold = 5
parameterRobot.p = 0.01
parameterRobot.indel = True
cluster.groupIndelNoisyKmers(noisyReads, parameterRobot, "fast")
def MSAResolverEpeatUnitTest(Nin, Lin, foldername = ""):
f2= logging.fmapfusedLoad('clusteredGroup2.csv')
G2 = logging.loadGraph(foldername+'basicMapping.csv', foldername+'seqMapping.txt', 'simple')
#checkCondensingBasic(G2, [G2[0]], "simple")
dummyParameters = logging.parameterObj()
dummyParameters.bridgingDepth = 5
dummyParameters.msaWidth = 20
G3 = bridgeResolve.resolveRepeats(f2,G2,dummyParameters)
#checkCondensingBasic(G3, [G3[0]], "MB")
N, G, L, p,snpRate, typeOfGen, detail = Nin, 10000,Lin, 0.015, 0.001 ,'m', "500-300-50"
motherGen, reads, noisyReads = logging.rawDataLoad(foldername+"UnitTest",G,N,L,"dn")
alignmentBridge.MSAresolve(f2, G3, noisyReads, snpRate,dummyParameters)
def clusterUnitTest2():
# Unit Test 1 : Generate 30 reads with 10 copies with noise , length being 20
motherGen, reads, noisyReads = logging.rawDataLoad("clusterReadsUnitTest",10,300,40,"dn")
parameterRobot = logging.parameterObj()
parameterRobot.N = 300
parameterRobot.L = 40
parameterRobot.G = 10000
parameterRobot.liid = 30
parameterRobot.K = 30
parameterRobot.threshold = 5
parameterRobot.p = 0.01
parameterRobot.indel = True
cluster.groupIndelNoisyKmers(noisyReads, parameterRobot, "fast")
def graphFormUnitTest(N, G, L, folderName=""):
print "dataGenUnitTest"
dummyParameters = logging.parameterObj()
dummyParameters.defaultFolder = folderName
dummyParameters.liid = 40
dummyParameters.K = 40
dummyParameters.threshold = 6
dummyParameters.p = 0.015
dummyParameters.indel = True
dummyParameters.N, dummyParameters.G, dummyParameters.L, dummyParameters.p, typeOfGen, detail = N, G, L, 0.015, 'm', "500-200-50"
# motherGen, reads, noisyReads = dataGen.generateData( typeOfGen,detail,dummyParameters)
motherGen, reads, noisyReads = logging.rawDataLoad(
dummyParameters.defaultFolder + "UnitTest", dummyParameters.G,
dummyParameters.N, dummyParameters.L, "dn")
returnfmapping = cluster.groupIndelNoisyKmers(noisyReads, dummyParameters,
"fast")
def segmentChopAndTestTest(Nin,G, Lin, foldername = "", bridgingDepth = 20, msaWidth = 20):
dummyParameters = logging.parameterObj()
dummyParameters.bridgingDepth = bridgingDepth
dummyParameters.msaWidth = msaWidth
dummyParameters.N, dummyParameters.G, dummyParameters.L, dummyParameters.p = Nin, G,Lin, 0.015
dummyParameters.indel = True
dummyParameters.defaultFolder = foldername
dummyParameters.threshold = 5
dummyParameters.liid = 48
snpRate, typeOfGen, detail = 0.001 ,'m', "500-200-50"
G,N,L = dummyParameters.G, dummyParameters.N, dummyParameters.L,
motherGen, reads, noisyReads = logging.rawDataLoad(foldername+"UnitTest",G,N,L, "dn")
#G3, f2 = interactiveNodeRemoval(folderName, "skip")
G3, f2 = interactiveNodeRemoval(folderName, "desiredOpt")
chopAndAlign(G3, f2 , noisyReads, motherGen, dummyParameters)
def MSAResolverEpeatUnitTest(Nin, Lin, foldername=""):
f2 = logging.fmapfusedLoad('clusteredGroup2.csv')
G2 = logging.loadGraph(foldername + 'basicMapping.csv',
foldername + 'seqMapping.txt', 'simple')
#checkCondensingBasic(G2, [G2[0]], "simple")
dummyParameters = logging.parameterObj()
dummyParameters.bridgingDepth = 5
dummyParameters.msaWidth = 20
G3 = bridgeResolve.resolveRepeats(f2, G2, dummyParameters)
#checkCondensingBasic(G3, [G3[0]], "MB")
N, G, L, p, snpRate, typeOfGen, detail = Nin, 10000, Lin, 0.015, 0.001, 'm', "500-300-50"
motherGen, reads, noisyReads = logging.rawDataLoad(foldername + "UnitTest",
G, N, L, "dn")
alignmentBridge.MSAresolve(f2, G3, noisyReads, snpRate, dummyParameters)
def ECUnitTest(Nin, G, Lin, foldername="", bridgingDepth=20, msaWidth=20):
dummyParameters = logging.parameterObj()
dummyParameters.bridgingDepth = bridgingDepth
dummyParameters.msaWidth = msaWidth
dummyParameters.N, dummyParameters.G, dummyParameters.L, dummyParameters.p = Nin, G, Lin, 0.015
dummyParameters.indel = True
dummyParameters.defaultFolder = foldername
dummyParameters.threshold = 5
dummyParameters.liid = 48
snpRate, typeOfGen, detail = 0.001, 'm', "500-200-50"
G, N, L = dummyParameters.G, dummyParameters.N, dummyParameters.L,
motherGen, reads, noisyReads = logging.rawDataLoad(foldername + "UnitTest",
G, N, L, "dn")
f2 = logging.fmapfusedLoad(foldername + 'clusteredGroup2.csv')
G2 = logging.loadGraph(foldername + 'basicMapping.csv',
foldername + 'seqMapping.txt', 'simple')
checkCondensingBasic(G2, [G2[0]], "simple")
G3 = bridgeResolve.resolveRepeats(f2, G2, dummyParameters)
checkCondensingBasic(G3, [G3[0]], "MB")
G4 = alignmentBridge.MSAresolve(f2, G3, noisyReads, snpRate,
dummyParameters)
#G4 = G3
#checkCondensingBasic(G4, [G4[0]], "MB")
recovSeq = eulerCycle.findEC(G4)
recovGen = readAns.reportRecovSeq(recovSeq, f2, noisyReads,
dummyParameters)
numMistakes, success = compare.subAlignCompare(recovGen, motherGen,
dummyParameters)
return numMistakes, success
def testChecking(folderName, N, L, G ):
#N, L, G = 1345, 1000, 50000
f2 = open(folderName+"rec.txt", 'r')
temp2 = f2.read()
recov = np.zeros(len(temp2), dtype = np.int32)
j =0
while j <len(temp2):
if temp2[j] != '-':
recov[j] = int(temp2[j])
j = j+1
else:
recov[j] = -1
j = j+2
f2.close()
dummyParameters = logging.parameterObj()
dummyParameters.defaultFolder = folderName
dummyParameters.G = G
motherGen, reads, noisyReads = logging.rawDataLoad(folderName+"UnitTest",G,N,L, 'd')
print len(recov) , len(motherGen)
numberMistakes, success = compare.subAlignCompare(recov, motherGen,dummyParameters)
def dataGenUnitTest2():
print "dataGenUnitTest2"
dummyParameters = logging.parameterObj()
dummyParameters.N, dummyParameters.G, dummyParameters.L, dummyParameters.p, typeOfGen, detail = 100, 10000,100, 0.01, 't', "500"
G,N,L = dummyParameters.G, dummyParameters.N, dummyParameters.L
motherGen, reads, noisyReads = logging.rawDataLoad("UnitTest",G,N,L, 'd')
print reads
print noisyReads
baseList = []
for indexN in range(N):
for indexL in range(L):
if reads[indexN][indexL] == noisyReads[indexN][indexL]:
baseList.append(0)
else:
baseList.append(1)
print "Error Probability", np.mean(baseList)
print "numberOfBasesGet ", len(baseList)
def dataGenUnitTest2():
print "dataGenUnitTest2"
dummyParameters = logging.parameterObj()
dummyParameters.N, dummyParameters.G, dummyParameters.L, dummyParameters.p, typeOfGen, detail = 100, 10000, 100, 0.01, 't', "500"
G, N, L = dummyParameters.G, dummyParameters.N, dummyParameters.L
motherGen, reads, noisyReads = logging.rawDataLoad("UnitTest", G, N, L,
'd')
print reads
print noisyReads
baseList = []
for indexN in range(N):
for indexL in range(L):
if reads[indexN][indexL] == noisyReads[indexN][indexL]:
baseList.append(0)
else:
baseList.append(1)
print "Error Probability", np.mean(baseList)
print "numberOfBasesGet ", len(baseList)
def segmentChopAndTestTest(Nin,
G,
Lin,
foldername="",
bridgingDepth=20,
msaWidth=20):
dummyParameters = logging.parameterObj()
dummyParameters.bridgingDepth = bridgingDepth
dummyParameters.msaWidth = msaWidth
dummyParameters.N, dummyParameters.G, dummyParameters.L, dummyParameters.p = Nin, G, Lin, 0.015
dummyParameters.indel = True
dummyParameters.defaultFolder = foldername
dummyParameters.threshold = 5
dummyParameters.liid = 48
snpRate, typeOfGen, detail = 0.001, 'm', "500-200-50"
G, N, L = dummyParameters.G, dummyParameters.N, dummyParameters.L,
motherGen, reads, noisyReads = logging.rawDataLoad(foldername + "UnitTest",
G, N, L, "dn")
#G3, f2 = interactiveNodeRemoval(folderName, "skip")
G3, f2 = interactiveNodeRemoval(folderName, "desiredOpt")
chopAndAlign(G3, f2, noisyReads, motherGen, dummyParameters)
def testChecking(folderName, N, L, G):
#N, L, G = 1345, 1000, 50000
f2 = open(folderName + "rec.txt", 'r')
temp2 = f2.read()
recov = np.zeros(len(temp2), dtype=np.int32)
j = 0
while j < len(temp2):
if temp2[j] != '-':
recov[j] = int(temp2[j])
j = j + 1
else:
recov[j] = -1
j = j + 2
f2.close()
dummyParameters = logging.parameterObj()
dummyParameters.defaultFolder = folderName
dummyParameters.G = G
motherGen, reads, noisyReads = logging.rawDataLoad(folderName + "UnitTest",
G, N, L, 'd')
print len(recov), len(motherGen)
numberMistakes, success = compare.subAlignCompare(recov, motherGen,
dummyParameters)