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"
def fetchSuccessor(folderName , mummerLink):
print "fetchSuccessor"
left_connect, right_connect = [], []
print "Direct greedy"
print "fetchSuccessor: Aligning non-contained contigs to themselves, output files are greedy*.delta"
numberOfContig, dataSet = IORobot.obtainLinkInfo(folderName, mummerLink, "noEmbed", "greedy")
# [next_item, overlap_length]
leftConnect = [[-1, -1] for i in range(numberOfContig)]
rightConnect = [[-1, -1] for i in range(numberOfContig)]
dataSet.sort(reverse=True, key=itemgetter(1))
print "fetchSuccessor: Finding best successors"
for key, items in groupby(dataSet, itemgetter(1)):
# if key == "Contig217_d":
# print "dddd"
maxVal = -1
myName = key
connectorName = ""
for eachsubitem in items:
if eachsubitem[0] > maxVal:
maxVal = eachsubitem[0]
connectorName = eachsubitem[2]
prefix = myName.split('_')
suffix = connectorName.split('_')
lengthOfOverlap = maxVal
if prefix[1] == 'p':
prefixContig = int(prefix[0][6:]) * 2
else:
prefixContig = int(prefix[0][6:]) * 2 + 1
if suffix[1] == 'p':
suffixContig = int(suffix[0][6:]) * 2
else:
suffixContig = int(suffix[0][6:]) * 2 + 1
assert(rightConnect[prefixContig][0] == -1)
rightConnect[prefixContig][0] = suffixContig
rightConnect[prefixContig][1] = lengthOfOverlap
dataSet.sort(reverse=True, key=itemgetter(2))
print "fetchSuccessor: Finding best predecessors"
for key, items in groupby(dataSet, itemgetter(2)):
maxVal = -1
myName = key
connectorName = ""
for eachsubitem in items:
if eachsubitem[0] > maxVal:
maxVal = eachsubitem[0]
connectorName = eachsubitem[1]
prefix = connectorName.split('_')
suffix = myName.split('_')
lengthOfOverlap = maxVal
if prefix[1] == 'p':
prefixContig = int(prefix[0][6:]) * 2
else:
prefixContig = int(prefix[0][6:]) * 2 + 1
if suffix[1] == 'p':
suffixContig = int(suffix[0][6:]) * 2
else:
suffixContig = int(suffix[0][6:]) * 2 + 1
assert(leftConnect[suffixContig][0] == -1)
leftConnect[suffixContig][0] = prefixContig
leftConnect[suffixContig][1] = lengthOfOverlap
print "fetchSuccessor: Outputting best successors to rightConnect.txt"
# ## Write to file:
f = open(folderName + 'rightConnect.txt', 'w')
for eachitem, dummyIndex in zip(rightConnect, range(len(rightConnect))):
f.write(str(dummyIndex) + ',' + str(eachitem[0]) + ',' + str(eachitem[1]) + '\n')
f.close()
print "fetchSuccessor: Outputting best predecessors to leftConnect.txt"
f = open(folderName + 'leftConnect.txt', 'w')
for eachitem, dummyIndex in zip(leftConnect, range(len(leftConnect))):
f.write(str(dummyIndex) + ',' + str(eachitem[0]) + ',' + str(eachitem[1]) + '\n')
f.close()
def fetchSuccessor(folderName , mummerLink):
print "fetchSuccessor"
left_connect, right_connect = [], []
print "Direct greedy"
numberOfContig, dataSet = IORobot.obtainLinkInfo(folderName, mummerLink, "noEmbed", "greedy")
# [next_item, overlap_length]
leftConnect = [[-1, -1] for i in range(numberOfContig)]
rightConnect = [[-1, -1] for i in range(numberOfContig)]
dataSet.sort(reverse=True, key=itemgetter(1))
for key, items in groupby(dataSet, itemgetter(1)):
# if key == "Contig217_d":
# print "dddd"
maxVal = -1
myName = key
connectorName = ""
for eachsubitem in items:
if eachsubitem[0] > maxVal:
maxVal = eachsubitem[0]
connectorName = eachsubitem[2]
prefix = myName.split('_')
suffix = connectorName.split('_')
lengthOfOverlap = maxVal
if prefix[1] == 'p':
prefixContig = int(prefix[0][6:]) * 2
else:
prefixContig = int(prefix[0][6:]) * 2 + 1
if suffix[1] == 'p':
suffixContig = int(suffix[0][6:]) * 2
else:
suffixContig = int(suffix[0][6:]) * 2 + 1
assert(rightConnect[prefixContig][0] == -1)
rightConnect[prefixContig][0] = suffixContig
rightConnect[prefixContig][1] = lengthOfOverlap
dataSet.sort(reverse=True, key=itemgetter(2))
for key, items in groupby(dataSet, itemgetter(2)):
maxVal = -1
myName = key
connectorName = ""
for eachsubitem in items:
if eachsubitem[0] > maxVal:
maxVal = eachsubitem[0]
connectorName = eachsubitem[1]
prefix = connectorName.split('_')
suffix = myName.split('_')
lengthOfOverlap = maxVal
if prefix[1] == 'p':
prefixContig = int(prefix[0][6:]) * 2
else:
prefixContig = int(prefix[0][6:]) * 2 + 1
if suffix[1] == 'p':
suffixContig = int(suffix[0][6:]) * 2
else:
suffixContig = int(suffix[0][6:]) * 2 + 1
assert(leftConnect[suffixContig][0] == -1)
leftConnect[suffixContig][0] = prefixContig
leftConnect[suffixContig][1] = lengthOfOverlap
# ## Write to file:
f = open(folderName + 'rightConnect.txt', 'w')
for eachitem, dummyIndex in zip(rightConnect, range(len(rightConnect))):
f.write(str(dummyIndex) + ',' + str(eachitem[0]) + ',' + str(eachitem[1]) + '\n')
f.close()
f = open(folderName + 'leftConnect.txt', 'w')
for eachitem, dummyIndex in zip(leftConnect, range(len(leftConnect))):
f.write(str(dummyIndex) + ',' + str(eachitem[0]) + ',' + str(eachitem[1]) + '\n')
f.close()
