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 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 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 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 fillInMissed(folderName, mummerLink, filerefname, filequeryname, fileoutname):
os.system("mv " + folderName + fileoutname + " " + folderName + filequeryname )
alignerRobot.useMummerAlignBatch(mummerLink, folderName, [[fileoutname+"fillmiss", filerefname, filequeryname, ""]], houseKeeper.globalParallel)
dataList = alignerRobot.extractMumData(folderName, fileoutname+"fillmissOut")
lenDic = obtainLength(folderName, filerefname)
### Check if there is any missing parts
# Format of the dataList : 1 765 | 11596 10822 | 765 775 | 84.25 | ref_NC_001133_ scf7180000000702"
dataList.sort(key = itemgetter(-2))
thres = 100
extraList = []
for key, items in groupby(dataList, itemgetter(-2)):
isFound = False
for eachitem in items:
if abs(int(eachitem[4]) - lenDic[key]) < thres:
isFound = True
break
if not isFound:
extraList.append(key)
### Fill in any missing items
referenceDic = loadContigsFromFile(folderName, filerefname)
queryDic = loadContigsFromFile(folderName, filequeryname)
ctgList = [referenceDic[eachitem] for eachitem in extraList] + [queryDic[eachitem] for eachitem in queryDic]
writeSegOut(ctgList, folderName, fileoutname)
print "fileoutname: len(extraList)",fileoutname, len(extraList), len(ctgList)
def formRelatedReadsFile(folderName, mummerLink):
# Find associated read and extract into a file associatedReads.fasta
# Input: contigs.fasta, cleaned_Reads.fasta
# Output: relatedReads.fasta
# ## Extract heads of the contigs
print ">formRelatedReadsFile"
f = open(folderName + "improved.fasta", 'r')
f2 = open(folderName + "improvedTrunc.fasta", 'w')
temp = f.readline()
tempContig = ""
thres = 400
runningIndex = 0
endThres = 10
while len(temp) > 0:
if temp[-1] == '\n':
temp = temp[0:-1]
if temp[0] == '>':
if len(tempContig) > 0:
IORobot.writeToFile(f2, runningIndex, tempContig[0:thres])
runningIndex = runningIndex + 1
IORobot.writeToFile(f2, runningIndex, tempContig[-thres:])
runningIndex = runningIndex + 1
IORobot.writeToFile(f2, runningIndex, houseKeeper.reverseComplement(tempContig[0:thres]))
runningIndex = runningIndex + 1
IORobot.writeToFile(f2, runningIndex, houseKeeper.reverseComplement(tempContig[-thres:]))
runningIndex = runningIndex + 1
tempContig = ""
else:
tempContig = tempContig + temp
temp = f.readline()
IORobot.writeToFile(f2, runningIndex, tempContig[0:thres])
runningIndex = runningIndex + 1
IORobot.writeToFile(f2, runningIndex, tempContig[-thres:])
runningIndex = runningIndex + 1
IORobot.writeToFile(f2, runningIndex, houseKeeper.reverseComplement(tempContig[0:thres]))
runningIndex = runningIndex + 1
IORobot.writeToFile(f2, runningIndex, houseKeeper.reverseComplement(tempContig[-thres:]))
runningIndex = runningIndex + 1
f2.close()
f.close()
# ## Write double stranded reads
IORobot.writeToFile_Double1(folderName, "improved.fasta", "improved_Double.fasta", "contig")
# writeToFile_Double1(folderName, "raw_reads.fasta", "raw_reads_Double.fasta","read")
# ## Apply MUMMER on them using cleanedReads against them
assoiatedReadIndex = []
nameList = []
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 + houseKeeper.globalReadName
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= "outGapFillRaw"+indexOfMum , "improvedTrunc.fasta", houseKeeper.globalReadName[0:-6] + ".part-" + indexOfMum + ".fasta", "fromMum" + indexOfMum
workerList.append([outputName, referenceName, queryName, specialName])
if True:
alignerRobot.useMummerAlignBatch(mummerLink, folderName, workerList, houseKeeper.globalParallel ,True)
# alignerRobot.useMummerAlign(mummerLink, folderName, "out", "improvedTrunc.fasta", "raw_reads.part-" + indexOfMum + ".fasta", True, "fromMum" + indexOfMum )
'''
command = mummerLink + "nucmer --maxmatch --nosimplify -p " + folderName + "out " + folderName + "improvedTrunc.fasta raw_reads.part-" + indexOfMum + ".fasta"
os.system(command)
command = mummerLink + "show-coords -r " + folderName + "out.delta > " + folderName + "fromMum" + indexOfMum
os.system(command)
'''
for dummyI in range(1, numberOfFiles + 1):
indexOfMum = ""
if dummyI < 10:
indexOfMum = "0" + str(dummyI)
else:
indexOfMum = str(dummyI)
f = open(folderName + "fromMum" + indexOfMum, 'r')
for i in range(6):
tmp = f.readline()
while len(tmp) > 0:
infoArr = tmp.split('|')
myArr = infoArr[-1].split('\t')
rdGpArr = infoArr[-1].split('\t')
contigName = rdGpArr[0].rstrip().lstrip()
readName = rdGpArr[1].rstrip().lstrip()
endSegArr = infoArr[0].split(" ")
pos = []
for eachitem in endSegArr:
if len(eachitem) > 0:
pos.append(int(eachitem))
startPos = pos[0]
endPos = pos[1]
if startPos < endThres and endPos > thres - endThres:
assoiatedReadIndex.append(myArr[1])
nameList.append([int(contigName.split('_')[1]), readName])
tmp = f.readline()
f.close()
nameList.sort()
assoiatedReadIndex.sort()
# print "assoiatedReadIndex", assoiatedReadIndex
ckIndex = 0
f = open(folderName + "associatedNames.txt", 'w')
oneItem = 0
keyFound = []
for key, items in groupby(assoiatedReadIndex):
countItem = 0
for eachitem in items:
countItem += 1
if countItem == 1:
oneItem += 1
else:
key = key.rstrip()
if not key in keyFound:
f.write(key + '\n')
keyFound.append(key)
ckIndex += 1
print "ckIndex,oneItem: ", ckIndex, oneItem
f.close()
fFilter = open(folderName + "associatedNames.txt", 'r')
fout = open(folderName + "associatedNames2.txt", 'w')
maxCount = 12000
mytmpDum = fFilter.readline()
i = 0
while i < maxCount and len(mytmpDum) > 0:
fout.write(mytmpDum)
mytmpDum = fFilter.readline()
i = i + 1
fout.close()
fFilter.close()
command = "perl -ne 'if(/^>(\S+)/){$c=$i{$1}}$c?print:chomp;$i{$_}=1 if @ARGV' " + folderName + "associatedNames2.txt " + folderName + houseKeeper.globalReadName +" > " + folderName + "relatedReads.fasta"
os.system(command)
IORobot.writeToFile_Double1(folderName, "relatedReads.fasta", "relatedReads_Double.fasta", "read")
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 obtainLinkInfo(folderName, mummerLink, inputFile, mummerFile):
thres = 5
minLen = 400
# thres = 10
# minLen = 200
writeToFile_Double1(folderName, inputFile + ".fasta",
inputFile + "_Double.fasta", "contig")
fmyFile = open(folderName + inputFile + "_Double.fasta", 'r')
fSmaller = open(folderName + inputFile + "_contigs_Double.fasta", 'w')
tmp = fmyFile.readline().rstrip()
maxSize = 50000
myName = ""
while len(tmp) > 0:
if tmp[0] == '>':
fSmaller.write(tmp + '\n')
myName = tmp[1:]
else:
component = tmp[0:min(len(tmp), maxSize)]
countComp = len(component)
fSmaller.write(component)
component = tmp[max(0, len(tmp) - maxSize):len(tmp)]
fSmaller.write(component)
countComp = countComp + len(component)
print "DebugName", myName, countComp
fSmaller.write('\n')
tmp = fmyFile.readline().rstrip()
fSmaller.close()
fmyFile.close()
if True:
alignerRobot.useMummerAlignBatch(mummerLink, folderName, [[
mummerFile, inputFile + "_contigs_Double.fasta",
inputFile + "_contigs_Double.fasta", ""
]], houseKeeper.globalParallel)
# alignerRobot.useMummerAlign(mummerLink, folderName, mummerFile, inputFile + "_contigs_Double.fasta", inputFile + "_contigs_Double.fasta")
lengthDic = obtainLength(folderName, inputFile + "_contigs_Double.fasta")
dataSetRaw = alignerRobot.extractMumData(folderName, mummerFile + "Out")
# ## Format [ helperStart, helperEnd , readStart, readEnd,matchLen1,matchLen2,percentMatch,helperName,readName]
dataSet = []
for eachitem in dataSetRaw:
helperStart, helperEnd, readStart, readEnd, matchLen1, matchLen2, percentMatch, helperName, readName = eachitem
detailHelper = helperName.split('_')
detailRead = readName.split('_')
if detailHelper[0] != detailRead[0] and helperName != readName and max(
matchLen1, matchLen2) > minLen and readStart < readEnd and min(
helperStart, readStart) < thres and min(
lengthDic[helperName] - helperEnd,
lengthDic[readName] - readEnd) + 1 < thres:
conditionForMatch = True
else:
conditionForMatch = False
if conditionForMatch:
if helperStart < thres:
dataSet.append((max(matchLen1,
matchLen2), readName, helperName))
dataSet.sort(reverse=True)
numberOfContig = len(lengthDic)
return numberOfContig, dataSet
def obtainLinkInfo(folderName, mummerLink, inputFile, mummerFile):
# minLen = 200
minLen = 400
if houseKeeper.globalRelaxThres == False:
thres = 5
elif houseKeeper.globalRelaxThres == True:
thres = 10
writeToFile_Double1(folderName, inputFile + ".fasta", inputFile + "_Double.fasta", "contig")
fmyFile = open(folderName + inputFile + "_Double.fasta", 'r')
fSmaller = open(folderName + inputFile + "_contigs_Double.fasta", 'w')
tmp = fmyFile.readline().rstrip()
maxSize = 50000
myName = ""
while len(tmp) > 0:
if tmp[0] == '>':
fSmaller.write(tmp + '\n')
myName = tmp[1:]
else:
component = tmp[0:min(len(tmp), maxSize)]
countComp = len(component)
fSmaller.write(component)
component = tmp[max(0, len(tmp) - maxSize):len(tmp)]
fSmaller.write(component)
countComp = countComp + len(component)
print "DebugName", myName, countComp
fSmaller.write('\n')
tmp = fmyFile.readline().rstrip()
fSmaller.close()
fmyFile.close()
if True:
alignerRobot.useMummerAlignBatch(mummerLink, folderName, [[mummerFile, inputFile + "_contigs_Double.fasta", inputFile + "_contigs_Double.fasta", ""]], houseKeeper.globalParallel )
# alignerRobot.useMummerAlign(mummerLink, folderName, mummerFile, inputFile + "_contigs_Double.fasta", inputFile + "_contigs_Double.fasta")
lengthDic = obtainLength(folderName, inputFile + "_contigs_Double.fasta")
dataSetRaw = alignerRobot.extractMumData(folderName, mummerFile + "Out")
# ## Format [ helperStart, helperEnd , readStart, readEnd,matchLen1,matchLen2,percentMatch,helperName,readName]
dataSet = []
for eachitem in dataSetRaw:
helperStart, helperEnd , readStart, readEnd, matchLen1, matchLen2, percentMatch, helperName, readName = eachitem
detailHelper = helperName.split('_')
detailRead = readName.split('_')
if detailHelper[0] != detailRead[0] and helperName != readName and max(matchLen1, matchLen2) > minLen and readStart < readEnd and min(helperStart, readStart) < thres and min(lengthDic[helperName] - helperEnd, lengthDic[readName] - readEnd) + 1 < thres:
conditionForMatch = True
else:
conditionForMatch = False
if conditionForMatch :
if helperStart < thres:
dataSet.append((max(matchLen1, matchLen2), readName, helperName))
dataSet.sort(reverse=True)
numberOfContig = len(lengthDic)
return numberOfContig, dataSet