def removeRedundantRefvsQuery(folderName, mummerLink, fileR, fileQ,
outputFileName):
thres = 10
if True:
alignerRobot.useMummerAlignBatch(mummerLink, folderName,
[["redundantRvsQ", fileR, fileQ, ""]],
houseKeeper.globalParallel)
dataList = alignerRobot.extractMumData(folderName, "redundantRvsQOut")
lenDicR = IORobot.obtainLength(folderName, fileR)
lenDicQ = IORobot.obtainLength(folderName, fileQ)
isRedundantList = []
for eachitem in dataList:
match1, match2, name1, name2 = eachitem[4], eachitem[5], eachitem[
7], eachitem[8]
l1, l2 = lenDicR[name1], lenDicQ[name2]
if abs(l2 - match2) < thres:
isRedundantList.append(name2)
#print lenDicQ
nonRedundantList = obtainComplement(lenDicQ, isRedundantList)
print nonRedundantList
IORobot.putListToFileO(folderName, fileQ, outputFileName, nonRedundantList)
os.system("cp " + folderName + "SC_n_tmp.fasta " + folderName +
"SC_n.fasta")
def removeRedundantRefvsQuery(folderName, mummerLink, fileR, fileQ, outputFileName):
thres = 10
if True:
alignerRobot.useMummerAlignBatch(
mummerLink, folderName, [["redundantRvsQ", fileR, fileQ, ""]], houseKeeper.globalParallel
)
dataList = alignerRobot.extractMumData(folderName, "redundantRvsQOut")
lenDicR = IORobot.obtainLength(folderName, fileR)
lenDicQ = IORobot.obtainLength(folderName, fileQ)
isRedundantList = []
for eachitem in dataList:
match1, match2, name1, name2 = eachitem[4], eachitem[5], eachitem[7], eachitem[8]
l1, l2 = lenDicR[name1], lenDicQ[name2]
if abs(l2 - match2) < thres:
isRedundantList.append(name2)
# print lenDicQ
nonRedundantList = obtainComplement(lenDicQ, isRedundantList)
print nonRedundantList
IORobot.putListToFileO(folderName, fileQ, outputFileName, nonRedundantList)
os.system("cp " + folderName + "SC_n_tmp.fasta " + folderName + "SC_n.fasta")
def removeEmbedded(folderName , mummerLink):
print "removeEmbedded"
thres = 10
os.system("sed -e 's/|//g' " + folderName + "contigs.fasta > " + folderName + "contigs2.fasta")
os.system("cp " + folderName + "contigs2.fasta " + folderName + "contigs.fasta")
if not os.path.isfile(folderName + "selfOut"):
alignerRobot.useMummerAlignBatch(mummerLink, folderName, [["self", "contigs.fasta", "contigs.fasta", ""]], houseKeeper.globalParallel )
# alignerRobot.useMummerAlign(mummerLink, folderName, "self", "contigs.fasta", "contigs.fasta")
# outputName, referenceName, queryName, specialName
lenDic = IORobot.obtainLength(folderName, 'contigs.fasta')
removeList = alignerRobot.extractMumDataAndRemove(folderName,"selfOut",lenDic,thres)
nameList = []
for eachitem in lenDic:
nameList.append(eachitem)
print len(nameList)
for eachitem in removeList:
if eachitem in nameList:
nameList.remove(eachitem)
print len(nameList)
IORobot.putListToFileO(folderName, "contigs.fasta", "noEmbed", nameList)
def removeRedundantWithFile(folderName, mummerLink, inputFilename,
mummerTmpName, outputFileName):
thres = 10
os.system("sed -e 's/|//g' " + folderName + inputFilename + ".fasta > " +
folderName + inputFilename + "2.fasta")
os.system("cp " + folderName + inputFilename + "2.fasta " + folderName +
inputFilename + ".fasta")
if True:
alignerRobot.useMummerAlignBatch(mummerLink, folderName, [[
mummerTmpName, inputFilename + ".fasta", inputFilename + ".fasta",
""
]], houseKeeper.globalParallel)
# alignerRobot.useMummerAlign(mummerLink, folderName, "self", "contigs.fasta", "contigs.fasta")
# outputName, referenceName, queryName, specialName
dataList = alignerRobot.extractMumData(folderName, mummerTmpName + "Out")
dataList = alignerRobot.transformCoor(dataList)
lenDic = IORobot.obtainLength(folderName, inputFilename + '.fasta')
removeList = []
shortEmbedClusterDic = {}
for eachitem in lenDic:
shortEmbedClusterDic[eachitem] = clusterElem(eachitem)
for eachitem in dataList:
match1, match2, name1, name2 = eachitem[4], eachitem[5], eachitem[
7], eachitem[8]
if name1 != name2:
l1, l2 = lenDic[name1], lenDic[name2]
if abs(l1 - match1) < thres and abs(l2 - match2) > thres:
removeList.append(name1)
elif abs(l1 - match1) > thres and abs(l2 - match2) < thres:
removeList.append(name2)
elif abs(l1 - match1) < thres and abs(l2 - match2) < thres:
print "Both shortembedd", eachitem
union(shortEmbedClusterDic[name1], shortEmbedClusterDic[name2])
nameList = obtainComplement(lenDic, removeList)
returnList = []
for eachitem in nameList:
if find(shortEmbedClusterDic[eachitem]).id == eachitem:
returnList.append(eachitem)
print "len(nameList), len(returnList)", len(nameList), len(returnList)
IORobot.putListToFileO(folderName, inputFilename + ".fasta",
outputFileName, returnList)
def removeRedundantWithFile(folderName, mummerLink, inputFilename, mummerTmpName, outputFileName):
thres = 10
os.system("sed -e 's/|//g' " + folderName + inputFilename + ".fasta > " + folderName + inputFilename + "2.fasta")
os.system("cp " + folderName + inputFilename + "2.fasta " + folderName + inputFilename + ".fasta")
if True:
alignerRobot.useMummerAlignBatch(
mummerLink,
folderName,
[[mummerTmpName, inputFilename + ".fasta", inputFilename + ".fasta", ""]],
houseKeeper.globalParallel,
)
# alignerRobot.useMummerAlign(mummerLink, folderName, "self", "contigs.fasta", "contigs.fasta")
# outputName, referenceName, queryName, specialName
dataList = alignerRobot.extractMumData(folderName, mummerTmpName + "Out")
dataList = alignerRobot.transformCoor(dataList)
lenDic = IORobot.obtainLength(folderName, inputFilename + ".fasta")
removeList = []
shortEmbedClusterDic = {}
for eachitem in lenDic:
shortEmbedClusterDic[eachitem] = clusterElem(eachitem)
for eachitem in dataList:
match1, match2, name1, name2 = eachitem[4], eachitem[5], eachitem[7], eachitem[8]
if name1 != name2:
l1, l2 = lenDic[name1], lenDic[name2]
if abs(l1 - match1) < thres and abs(l2 - match2) > thres:
removeList.append(name1)
elif abs(l1 - match1) > thres and abs(l2 - match2) < thres:
removeList.append(name2)
elif abs(l1 - match1) < thres and abs(l2 - match2) < thres:
print "Both shortembedd", eachitem
union(shortEmbedClusterDic[name1], shortEmbedClusterDic[name2])
nameList = obtainComplement(lenDic, removeList)
returnList = []
for eachitem in nameList:
if find(shortEmbedClusterDic[eachitem]).id == eachitem:
returnList.append(eachitem)
print "len(nameList), len(returnList)", len(nameList), len(returnList)
IORobot.putListToFileO(folderName, inputFilename + ".fasta", outputFileName, returnList)
def removeEmbedded(folderName, mummerLink):
print "removeEmbedded"
thres = 10
os.system("sed -e 's/|//g' " + folderName + "contigs.fasta > " +
folderName + "contigs2.fasta")
os.system("cp " + folderName + "contigs2.fasta " + folderName +
"contigs.fasta")
if True:
print "removeEmbedded: Aligning contigs.fasta to contigs.fasta, outputs are self*.delta"
alignerRobot.useMummerAlignBatch(
mummerLink, folderName,
[["self", "contigs.fasta", "contigs.fasta", ""]],
houseKeeper.globalParallel)
# alignerRobot.useMummerAlign(mummerLink, folderName, "self", "contigs.fasta", "contigs.fasta")
# outputName, referenceName, queryName, specialName
print "removeEmbedded: Extracting MUMmer data from delta files to selfOut"
dataList = alignerRobot.extractMumData(folderName, "selfOut")
dataList = alignerRobot.transformCoor(dataList)
lenDic = IORobot.obtainLength(folderName, 'contigs.fasta')
removeList = []
for eachitem in dataList:
match1, match2, name1, name2 = eachitem[4], eachitem[5], eachitem[
7], eachitem[8]
if name1 != name2:
l1, l2 = lenDic[name1], lenDic[name2]
if abs(l1 - match1) < thres and abs(l2 - match2) > thres:
removeList.append(name1)
elif abs(l1 - match1) > thres and abs(l2 - match2) < thres:
removeList.append(name2)
elif abs(l1 - match1) < thres and abs(l2 - match2) < thres:
print "Both shortembedd", eachitem
nameList = []
for eachitem in lenDic:
nameList.append(eachitem)
print len(nameList)
for eachitem in removeList:
if eachitem in nameList:
nameList.remove(eachitem)
print len(nameList)
print "removeEmbedded: Outputting non-contained contigs to noEmbed.fasta"
IORobot.putListToFileO(folderName, "contigs.fasta", "noEmbed", nameList)
def removeEmbedded(folderName, mummerLink):
print "removeEmbedded"
thres = 10
command = r'''perl -pe 's/>[^\$]*$/">Seg" . ++$n ."\n"/ge' ''' + folderName + "raw_reads.fasta > " + folderName + houseKeeper.globalReadName
os.system(command)
command = r'''perl -pe 's/>[^\$]*$/">Seg" . ++$n ."\n"/ge' ''' + folderName + "contigs.fasta > " + folderName + houseKeeper.globalContigName
os.system(command)
if True:
alignerRobot.useMummerAlignBatch(mummerLink, folderName, [[
"self", houseKeeper.globalContigName, houseKeeper.globalContigName,
""
]], houseKeeper.globalParallel)
# alignerRobot.useMummerAlign(mummerLink, folderName, "self", "contigs.fasta", "contigs.fasta")
# outputName, referenceName, queryName, specialName
dataList = alignerRobot.extractMumData(folderName, "selfOut")
dataList = alignerRobot.transformCoor(dataList)
lenDic = IORobot.obtainLength(folderName, houseKeeper.globalContigName)
removeList = []
for eachitem in dataList:
match1, match2, name1, name2 = eachitem[4], eachitem[5], eachitem[
7], eachitem[8]
if name1 != name2:
l1, l2 = lenDic[name1], lenDic[name2]
if abs(l1 - match1) < thres and abs(l2 - match2) > thres:
removeList.append(name1)
elif abs(l1 - match1) > thres and abs(l2 - match2) < thres:
removeList.append(name2)
elif abs(l1 - match1) < thres and abs(l2 - match2) < thres:
print "Both shortembedd", eachitem
nameList = []
for eachitem in lenDic:
nameList.append(eachitem)
print len(nameList)
for eachitem in removeList:
if eachitem in nameList:
nameList.remove(eachitem)
print len(nameList)
IORobot.putListToFileO(folderName, houseKeeper.globalContigName, "noEmbed",
nameList)
def removeEmbedded(folderName , mummerLink):
print "removeEmbedded"
thres = 10
command= r'''perl -pe 's/>[^\$]*$/">Seg" . ++$n ."\n"/ge' ''' + folderName + "raw_reads.fasta > " + folderName + houseKeeper.globalReadName
os.system(command)
command= r'''perl -pe 's/>[^\$]*$/">Seg" . ++$n ."\n"/ge' ''' + folderName + "contigs.fasta > " + folderName + houseKeeper.globalContigName
os.system(command)
if True:
alignerRobot.useMummerAlignBatch(mummerLink, folderName, [["self", houseKeeper.globalContigName, houseKeeper.globalContigName, ""]], houseKeeper.globalParallel)
# alignerRobot.useMummerAlign(mummerLink, folderName, "self", "contigs.fasta", "contigs.fasta")
# outputName, referenceName, queryName, specialName
dataList = alignerRobot.extractMumData(folderName, "selfOut")
dataList = alignerRobot.transformCoor(dataList)
lenDic = IORobot.obtainLength(folderName, houseKeeper.globalContigName)
removeList = []
for eachitem in dataList:
match1, match2, name1, name2 = eachitem[4], eachitem[5], eachitem[7], eachitem[8]
if name1 != name2:
l1, l2 = lenDic[name1], lenDic[name2]
if abs(l1 - match1) < thres and abs(l2 - match2) > thres:
removeList.append(name1)
elif abs(l1 - match1) > thres and abs(l2 - match2) < thres:
removeList.append(name2)
elif abs(l1 - match1) < thres and abs(l2 - match2) < thres:
print "Both shortembedd", eachitem
nameList = []
for eachitem in lenDic:
nameList.append(eachitem)
print len(nameList)
for eachitem in removeList:
if eachitem in nameList:
nameList.remove(eachitem)
print len(nameList)
IORobot.putListToFileO(folderName, houseKeeper.globalContigName, "noEmbed", nameList)
def runningTestSet(self ,myFolderName, ctexpected):
print "Integration test on FinisherSC: " + myFolderName
self.sourceFolder = myFolderName
os.system("mkdir " + self.testingFolder)
for eachitem in self.listOfFiles:
os.system("cp "+ self.sourceFolder + eachitem + " " +self.testingFolder)
os.system("python finisherSC.py -par 4 "+ self.testingFolder + " "+ self.mummerPath)
lenDic = IORobot.obtainLength(self.testingFolder, "/improved3.fasta")
print lenDic
assert(len(lenDic) == ctexpected)
os.system("rm -rf "+ self.testingFolder)
def runningTestSet(self ,myFolderName, ctexpected):
print "Integration test on FinisherSC: " + myFolderName
self.sourceFolder = myFolderName
os.system("mkdir " + self.testingFolder)
for eachitem in self.listOfFiles:
os.system("cp "+ self.sourceFolder + eachitem + " " +self.testingFolder)
os.system("python finisherSC.py -par 4 "+ self.testingFolder + " "+ self.mummerPath)
lenDic = IORobot.obtainLength(self.testingFolder, "/improved3.fasta")
print lenDic
assert(len(lenDic) == ctexpected)
os.system("rm -rf "+ self.testingFolder)
def removeEmbedded(folderName, mummerLink):
print "removeEmbedded"
thres = 10
os.system("sed -e 's/|//g' " + folderName + "contigs.fasta > " + folderName + "contigs2.fasta")
os.system("cp " + folderName + "contigs2.fasta " + folderName + "contigs.fasta")
if True:
print "removeEmbedded: Aligning contigs.fasta to contigs.fasta, outputs are self*.delta"
alignerRobot.useMummerAlignBatch(
mummerLink, folderName, [["self", "contigs.fasta", "contigs.fasta", ""]], houseKeeper.globalParallel
)
# alignerRobot.useMummerAlign(mummerLink, folderName, "self", "contigs.fasta", "contigs.fasta")
# outputName, referenceName, queryName, specialName
print "removeEmbedded: Extracting MUMmer data from delta files to selfOut"
dataList = alignerRobot.extractMumData(folderName, "selfOut")
dataList = alignerRobot.transformCoor(dataList)
lenDic = IORobot.obtainLength(folderName, "contigs.fasta")
removeList = []
for eachitem in dataList:
match1, match2, name1, name2 = eachitem[4], eachitem[5], eachitem[7], eachitem[8]
if name1 != name2:
l1, l2 = lenDic[name1], lenDic[name2]
if abs(l1 - match1) < thres and abs(l2 - match2) > thres:
removeList.append(name1)
elif abs(l1 - match1) > thres and abs(l2 - match2) < thres:
removeList.append(name2)
elif abs(l1 - match1) < thres and abs(l2 - match2) < thres:
print "Both shortembedd", eachitem
nameList = []
for eachitem in lenDic:
nameList.append(eachitem)
print len(nameList)
for eachitem in removeList:
if eachitem in nameList:
nameList.remove(eachitem)
print len(nameList)
print "removeEmbedded: Outputting non-contained contigs to noEmbed.fasta"
IORobot.putListToFileO(folderName, "contigs.fasta", "noEmbed", nameList)
def observeOverlap(folderName):
dataList = alignerRobot.extractMumData(folderName, "selfOut")
dataList = alignerRobot.transformCoor(dataList)
lenDic = IORobot.obtainLength(folderName, 'contigs.fasta')
matchThres = 10000
nonMatchThres = 500
count = 0
newDataList = []
for eachitem in dataList:
name1, name2 = eachitem[-2], eachitem[-1]
matchLen1, matchLen2 = eachitem[4], eachitem[5]
start1, end1, start2, end2 = eachitem[0], eachitem[1], eachitem[
2], eachitem[3]
# if name1!= name2 and min(lenDic[name1] - end1, lenDic[name2] - end2 ) > nonMatchThres \
# and min(start1, start2) > nonMatchThres \
if name1!= name2 and ( min(lenDic[name1] - end1, lenDic[name2] - end2 ) > nonMatchThres \
or min(start1, start2) > nonMatchThres ) \
and matchLen1> matchThres:
print "eachitem ", eachitem, lenDic[name1], lenDic[name2]
count = count + 1
newDataList.append(eachitem)
print "Count: " + str(count)
blkDic = getBreakPointFromDataList(folderName, newDataList)
LCList = IORobot.loadContigsFromFile(folderName, "contigs.fasta")
contigList = []
for eachcontig in LCList:
#print eachcontig
if not eachcontig in blkDic:
contigList = contigList + [LCList[eachcontig]]
else:
contigList = contigList + tmpBreakAcBkPts(LCList[eachcontig],
blkDic[eachcontig])
print "len(contigList)", len(contigList)
IORobot.writeSegOut(contigList, folderName, "breakChains.fasta")
def getBreakPointFromDataList(folderName, dataList):
g = 1000
blkDic = {}
dataList.sort(key=itemgetter(-2))
lenDic = IORobot.obtainLength(folderName, "contigs.fasta")
json_data = open(folderName + "modifiedOutliners.json", 'r')
breakPtsDic = json.load(json_data)
sep = 5000
for key, items in groupby(dataList, itemgetter(-2)):
contigName = key
newList = []
for eachitem in items:
newList.append([eachitem[0], eachitem[1]])
newList.sort()
bktmp = [0]
if newList[0][0] > g:
if withinBound(sep, breakPtsDic[contigName], newList[0][0]):
bktmp.append(newList[0][0])
#bktmp.append(newList[0][0])
for i in range(len(newList) - 1):
if newList[i + 1][0] > newList[i][1] + g:
if withinBound(sep, breakPtsDic[contigName],
newList[i + 1][0]):
bktmp.append(newList[i + 1][0])
bktmp.append(lenDic[contigName])
blkDic[contigName] = bktmp
print "contigName: " + contigName
print "bktmp:", bktmp
print "breakPtsDic[contigName]", breakPtsDic[contigName]
return blkDic
def extractEdgeSet(folderName, mummerLink, option="nopolish"):
# Tasks: reconstruct the string graph
# Input : relatedReads_Double.fasta, conig_Double.fasta
# Intermediate files: fromMum_overlap , fromMum_overlap
# Output: connectivity of eachNode: InList, OutList [critical]
# connectivity of eachNode: arrow representation with size [optional]
# ## Perform MUMMER alignment
print ">Extract Edge set"
contigOnlyLengthDic = IORobot.obtainLength(folderName, "improved.fasta")
# print lengthDic
lengthDic = IORobot.findContigLength(folderName, "improved")
numberOfContig = len(contigOnlyLengthDic)*2
K = 400
thres = 5
# ## Apply MUMMER on them using cleanedReads against them
IORobot.truncateEndOfContigs(folderName, "improved_Double.fasta", "smaller_improvedContig.fasta", 25000, lengthDic)
dataSet = []
numberOfFiles = max(20, houseKeeper.globalParallel)
if True:
bindir = os.path.abspath(os.path.dirname(sys.argv[0]))
command = bindir + "/fasta-splitter.pl --n-parts " + str(numberOfFiles) + " " + folderName + "relatedReads_Double.fasta"
os.system(command)
workerList = []
for dummyI in range(1, numberOfFiles + 1):
indexOfMum = ""
if dummyI < 10:
indexOfMum = "0" + str(dummyI)
else:
indexOfMum = str(dummyI)
outputName, referenceName, queryName, specialName= "outGapFillRefine"+indexOfMum , "smaller_improvedContig.fasta", "relatedReads_Double.part-" + indexOfMum + ".fasta", "fromMumRefine" + indexOfMum
workerList.append([outputName, referenceName, queryName, specialName])
if True:
alignerRobot.useMummerAlignBatch(mummerLink, folderName, workerList, houseKeeper.globalParallel ,True)
# alignerRobot.useMummerAlign(mummerLink, folderName, "outRefine", "smaller_improvedContig.fasta", "relatedReads_Double.part-" + indexOfMum + ".fasta", True, "fromMumRefine" + indexOfMum)
for dummyI in range(1, numberOfFiles + 1):
tmpSet = IORobot.obtainLinkInfoReadContig(dummyI, mummerLink, folderName,thres, lengthDic, K)
dataSet = dataSet + tmpSet
# ## repeat aware
usableJunction = loadOpenList(folderName)
dataSet, blockedSet = filterRepeatEnd(dataSet, usableJunction)
# ## repeat aware end
dataSet.sort()
matchPair = formMatchPairFromReadInfo(dataSet)
# Bug fix on repeat detection from reads alone
matchPair = filterRepeatPair(matchPair)
# end bug fix
# print matchPair
bestMatchPair = []
for key, items in groupby(matchPair, itemgetter(0, 1)):
maxvalue = -1
maxLenPair = []
for eachitem in items:
if eachitem[2] > maxvalue:
maxvalue = eachitem[2]
maxLenPair = [eachitem[3], eachitem[4], eachitem[5]]
bestMatchPair.append([key[0], key[1], maxvalue, maxLenPair[0], maxLenPair[1], maxLenPair[2]])
contigList, leftConnect, rightConnect, rawReadList = formbestpair(bestMatchPair,numberOfContig)
print "contigList", contigList
writeContigReadCombine(blockedSet, dataSet, folderName, rawReadList, numberOfContig, contigList, leftConnect, option, rightConnect, mummerLink)
def xPhased(folderName , mummerLink):
# ## Repeat resolution [Proxy for MB]
# 1. Re-form the contig string graph with ALL connections from contigs only V
# 2. Log down the reads and associated blocked contigs V
# 3. Use reads to connect;
# 4. Transform graph by identifying 1 successor/predecessor case ; Condense(important);
# 5. Read out contigs
numberOfContig, dataSet = IORobot.obtainLinkInfo(folderName, mummerLink, "improved2", "mb")
lenDic = IORobot.obtainLength(folderName, "improved2_Double.fasta")
confidenLenThres = 0
G = graphLib.seqGraph(numberOfContig)
extraEdges = loadEdgeFromBlockedReads(folderName)
for eachitem in dataSet:
# print eachitem
wt, myin, myout = eachitem
myInData = myin[6:].split('_')
myOutData = myout[6:].split('_')
if myInData[1] == 'p':
offsetin = 0
else:
offsetin = 1
if myOutData[1] == 'p':
offsetout = 0
else:
offsetout = 1
i = int(myInData[0]) * 2 + offsetin
j = int(myOutData[0]) * 2 + offsetout
ck = False
for eachedge in extraEdges:
mystart, myend, len1, len2 = eachedge[0], eachedge[1], eachedge[2] , eachedge[3]
if [i, j] == [mystart, myend] and min(len1, len2) >= wt and lenDic[myin] >= confidenLenThres and lenDic[myout] >= confidenLenThres:
ck = True
if ck:
G.insertEdge(i, j, wt)
# G.reportEdge()
G.MBResolve()
G.reportEdge()
G.saveToFile(folderName, "condensedGraphMB.txt")
graphFileName = "condensedGraphMB.txt"
contigFile = "improved2_Double.fasta"
outContigFile = "improved3.fasta"
outOpenList = "openZoneMB.txt"
IORobot.readContigOut(folderName, mummerLink, graphFileName, contigFile, outContigFile, outOpenList)
# ## Repeat resolution [Proxy for phasing step]
# 6. Find out the repeat region by MSA
# 7. Find out the location of SNPs and extend across repeat
# [short cut : use contig creator : your job here is to get data into the correct formats]
print "xPhased"
def xPhased(folderName , mummerLink):
# ## Repeat resolution [Proxy for MB]
# 1. Re-form the contig string graph with ALL connections from contigs only V
# 2. Log down the reads and associated blocked contigs V
# 3. Use reads to connect;
# 4. Transform graph by identifying 1 successor/predecessor case ; Condense(important);
# 5. Read out contigs
print "xPhased: Aligning improved2.fasta against itself, outputting to mb*.delta"
numberOfContig, dataSet = IORobot.obtainLinkInfo(folderName, mummerLink, "improved2", "mb")
lenDic = IORobot.obtainLength(folderName, "improved2_Double.fasta")
confidenLenThres = 0
print "xPhased: Building seqGraph"
G = graphLib.seqGraph(numberOfContig)
extraEdges = loadEdgeFromBlockedReads(folderName)
for eachitem in dataSet:
# print eachitem
wt, myin, myout = eachitem
myInData = myin[6:].split('_')
myOutData = myout[6:].split('_')
if myInData[1] == 'p':
offsetin = 0
else:
offsetin = 1
if myOutData[1] == 'p':
offsetout = 0
else:
offsetout = 1
i = int(myInData[0]) * 2 + offsetin
j = int(myOutData[0]) * 2 + offsetout
ck = False
for eachedge in extraEdges:
mystart, myend, len1, len2 = eachedge[0], eachedge[1], eachedge[2] , eachedge[3]
if [i, j] == [mystart, myend] and min(len1, len2) >= wt and lenDic[myin] >= confidenLenThres and lenDic[myout] >= confidenLenThres:
ck = True
if ck:
G.insertEdge(i, j, wt)
# G.reportEdge()
G.MBResolve()
G.reportEdge()
print "xPhased: Saving condensed seqGraph to condensedGraphMB.txt"
G.saveToFile(folderName, "condensedGraphMB.txt")
graphFileName = "condensedGraphMB.txt"
contigFile = "improved2_Double.fasta"
outContigFile = "improved3.fasta"
outOpenList = "openZoneMB.txt"
print "xPhased: Outputting improved contigs from condensed seqGraph to improved3.fasta"
IORobot.readContigOut(folderName, mummerLink, graphFileName, contigFile, outContigFile, outOpenList)
# ## Repeat resolution [Proxy for phasing step]
# 6. Find out the repeat region by MSA
# 7. Find out the location of SNPs and extend across repeat
# [short cut : use contig creator : your job here is to get data into the correct formats]
print "xPhased"
