def abunSplit(folderName, mummerLink, myCountDic):
"""
Input : repeatSpecification.txt , myCountDic.json, improved3.fasta, raw_reads.fasta
Output : abunsplit.fasta
Algorithm :
1. Load data from various sources [various json files]
2. For each repeat interior:
a) identify the abundances associated with in/out contigs
b) perform a split and record the split
3. Use split results to generate contigs [may already exist in newPhasing.py ]
a) use a graph to capture the split results
b) use reads to fill in any gaps
c) read out the contigs
"""
json_data = open(folderName + "phaseRepeat.txt", "r")
repeatPairs = json.load(json_data)
repeatPairs = obtainNonEmpty(repeatPairs)
N1 = len(myCountDic) * 2
G = graphLib.seqGraph(N1)
for eachitem in repeatPairs:
inList, outList = eachitem[0], eachitem[1]
resolvedList = determindMatch(inList, outList, myCountDic)
addEdges(G, resolvedList)
G.condense()
IORobot.extractGraphToContigs(G, folderName, mummerLink, "abun.fasta", "improved3_Double.fasta")
def abunSplit(folderName, mummerLink, myCountDic):
'''
Input : repeatSpecification.txt , myCountDic.json, improved3.fasta, raw_reads.fasta
Output : abunsplit.fasta
Algorithm :
1. Load data from various sources [various json files]
2. For each repeat interior:
a) identify the abundances associated with in/out contigs
b) perform a split and record the split
3. Use split results to generate contigs [may already exist in newPhasing.py ]
a) use a graph to capture the split results
b) use reads to fill in any gaps
c) read out the contigs
'''
json_data = open(folderName + "phaseRepeat.txt", 'r')
repeatPairs = json.load(json_data)
repeatPairs = obtainNonEmpty(repeatPairs)
N1 = len(myCountDic)*2
G = graphLib.seqGraph(N1)
for eachitem in repeatPairs:
inList, outList = eachitem[0], eachitem[1]
resolvedList = determindMatch(inList, outList, myCountDic)
addEdges(G, resolvedList)
G.condense()
IORobot.extractGraphToContigs(G, folderName, mummerLink, "abun.fasta", "improved3_Double.fasta")
def abunSplit(folderName, mummerLink, myCountDic, contigReadGraph,
contigFilename, readsetFilename):
'''
Input : repeatSpecification.txt , myCountDic.json, improved3.fasta, raw_reads.fasta
Output : abunsplit.fasta
Algorithm :
1. Load data from various sources [various json files]
2. For each repeat interior:
a) identify the abundances associated with in/out contigs
b) perform a split and record the split
3. Use split results to generate contigs [may already exist in newPhasing.py ]
a) use a graph to capture the split results
b) use reads to fill in any gaps
c) read out the contigs
'''
json_data = open(folderName + "phaseRepeat.txt", 'r')
repeatPairs = json.load(json_data)
repeatPairs = obtainNonEmpty(repeatPairs)
N1 = len(myCountDic) * 2
print "N1", N1
G = graphLib.seqGraph(N1)
gapContentLookUpList = []
for eachitem in repeatPairs:
inList, outList = eachitem[0], eachitem[1]
resolvedList = determindMatch(inList, outList, myCountDic, folderName,
contigReadGraph, N1)
print "resolvedList", resolvedList
gapContentLookUpList += generateGapContentLookup(
folderName, mummerLink, resolvedList, contigReadGraph,
contigFilename, readsetFilename)
addEdges(G, resolvedList)
gapContentLookUpDic = {}
gapContentLookUpList.sort()
for eachitem in gapContentLookUpList:
gapContentLookUpDic[str(eachitem[0]) + "_" + str(eachitem[1])] = [
eachitem[2], eachitem[3], eachitem[4]
]
print eachitem[2:4], len(eachitem[4])
# some how change ASplitter here by appending necessary information
G.condense()
IORobot.extractGraphToContigs(G, folderName, mummerLink, "abun.fasta",
contigFilename + "_Double.fasta",
gapContentLookUpDic)
def abunSplit(folderName, mummerLink, myCountDic, contigReadGraph, contigFilename, readsetFilename):
"""
Input : repeatSpecification.txt , myCountDic.json, improved3.fasta, raw_reads.fasta
Output : abunsplit.fasta
Algorithm :
1. Load data from various sources [various json files]
2. For each repeat interior:
a) identify the abundances associated with in/out contigs
b) perform a split and record the split
3. Use split results to generate contigs [may already exist in newPhasing.py ]
a) use a graph to capture the split results
b) use reads to fill in any gaps
c) read out the contigs
"""
json_data = open(folderName + "phaseRepeat.txt", "r")
repeatPairs = json.load(json_data)
repeatPairs = obtainNonEmpty(repeatPairs)
N1 = len(myCountDic) * 2
print "N1", N1
G = graphLib.seqGraph(N1)
gapContentLookUpList = []
for eachitem in repeatPairs:
inList, outList = eachitem[0], eachitem[1]
resolvedList = determindMatch(inList, outList, myCountDic, folderName, contigReadGraph, N1)
print "resolvedList", resolvedList
gapContentLookUpList += generateGapContentLookup(
folderName, mummerLink, resolvedList, contigReadGraph, contigFilename, readsetFilename
)
addEdges(G, resolvedList)
gapContentLookUpDic = {}
gapContentLookUpList.sort()
for eachitem in gapContentLookUpList:
gapContentLookUpDic[str(eachitem[0]) + "_" + str(eachitem[1])] = [eachitem[2], eachitem[3], eachitem[4]]
print eachitem[2:4], len(eachitem[4])
# some how change ASplitter here by appending necessary information
G.condense()
IORobot.extractGraphToContigs(
G, folderName, mummerLink, "abun.fasta", contigFilename + "_Double.fasta", gapContentLookUpDic
)
def readContigForAbunSplit(folderName, mummerLink, contigFilename,
readsetFilename, N1, contigReadGraph):
json_data = open(folderName + "mapDummyToRealDic.json", 'r')
mapDummyToRealDic = json.load(json_data)
G = []
G = graphLib.seqGraph(0)
G.loadFromFile(folderName, "xResolvedGraph")
gapContentLookUpDic = {}
furtherGapList = []
for i in range(N1):
if len(G.graphNodesList[i].nodeIndexList) > 1:
for j in range(len(G.graphNodesList[i].nodeIndexList) - 1):
bk, fwd = G.graphNodesList[i].nodeIndexList[
j], G.graphNodesList[i].nodeIndexList[j + 1]
key = str(bk) + "_" + str(fwd)
if not key in gapContentLookUpDic:
furtherGapList.append([bk, fwd])
with open(folderName + "furtherGapList.json", 'w') as f:
json.dump(furtherGapList, f)
furtherGapContentLookUpList = generateGapContentLookup(
folderName, mummerLink, furtherGapList, contigReadGraph,
contigFilename, readsetFilename, mapDummyToRealDic)
for eachitem in furtherGapContentLookUpList:
gapContentLookUpDic[str(eachitem[0]) + "_" + str(eachitem[1])] = [
eachitem[2], eachitem[3], eachitem[4]
]
print eachitem[2:4], len(eachitem[4])
#segLookUp = IORobot.readContigsFromFile(folderName, "LC_n_Double.fasta")
print "Final step: really hacking a file"
os.system("cp " + folderName + contigFilename + "_Double.fasta " +
folderName + "tmpWithDummy.fasta")
contigList = IORobot.readContigsFromFile(folderName,
contigFilename + "_Double.fasta")
IORobot.extractGraphToContigs(G, folderName, mummerLink, "abunPre.fasta",
"tmpWithDummy.fasta", gapContentLookUpDic,
mapDummyToRealDic)
if True:
nonRedundantResolver.removeRedundantWithFile(folderName, mummerLink,
"abunPre", "abunMum",
"abun")
def readContigForAbunSplit(folderName, mummerLink, contigFilename, readsetFilename, N1, contigReadGraph):
json_data = open(folderName + "mapDummyToRealDic.json", "r")
mapDummyToRealDic = json.load(json_data)
G = []
G = graphLib.seqGraph(0)
G.loadFromFile(folderName, "xResolvedGraph")
gapContentLookUpDic = {}
furtherGapList = []
for i in range(N1):
if len(G.graphNodesList[i].nodeIndexList) > 1:
for j in range(len(G.graphNodesList[i].nodeIndexList) - 1):
bk, fwd = G.graphNodesList[i].nodeIndexList[j], G.graphNodesList[i].nodeIndexList[j + 1]
key = str(bk) + "_" + str(fwd)
if not key in gapContentLookUpDic:
furtherGapList.append([bk, fwd])
with open(folderName + "furtherGapList.json", "w") as f:
json.dump(furtherGapList, f)
furtherGapContentLookUpList = generateGapContentLookup(
folderName, mummerLink, furtherGapList, contigReadGraph, contigFilename, readsetFilename, mapDummyToRealDic
)
for eachitem in furtherGapContentLookUpList:
gapContentLookUpDic[str(eachitem[0]) + "_" + str(eachitem[1])] = [eachitem[2], eachitem[3], eachitem[4]]
print eachitem[2:4], len(eachitem[4])
# segLookUp = IORobot.readContigsFromFile(folderName, "LC_n_Double.fasta")
print "Final step: really hacking a file"
os.system("cp " + folderName + contigFilename + "_Double.fasta " + folderName + "tmpWithDummy.fasta")
contigList = IORobot.readContigsFromFile(folderName, contigFilename + "_Double.fasta")
IORobot.extractGraphToContigs(
G, folderName, mummerLink, "abunPre.fasta", "tmpWithDummy.fasta", gapContentLookUpDic, mapDummyToRealDic
)
if True:
nonRedundantResolver.removeRedundantWithFile(folderName, mummerLink, "abunPre", "abunMum", "abun")
def abunSplitWithXResolve(folderName, mummerLink, myCountDic, contigReadGraph, contigFilename, readsetFilename):
N1 = len(myCountDic) * 2
print "N1", N1
# Debug
G = graphLib.seqGraph(0)
G.loadFromFile(folderName, contigReadGraph)
adj = [[] for i in range(N1)]
for i in range(N1):
adj[i] = abunGraphLib.findAllReachable(i, N1, G)
Gnew = graphLib.seqGraph(N1)
for i in range(N1):
for j in adj[i]:
Gnew.insertEdge(i, j, 1)
Gnew.reportEdge()
# End Debug
if False:
json_data = open(folderName + "phaseRepeat.txt", "r")
repeatPairs = json.load(json_data)
repeatPairs = obtainNonEmpty(repeatPairs)
biResolvedCombineList = []
for eachitem in repeatPairs:
inList, outList = eachitem[0], eachitem[1]
resolvedList = determindMatch(inList, outList, myCountDic, folderName, contigReadGraph, N1)
biResolvedCombineList += resolvedList
### Xnode repeatResolution
xResolvedList, mapDummyToRealDic = xNodeResolving(folderName, contigReadGraph)
### Combine resolution
resolvedList = xResolvedList + biResolvedCombineList
resolvedList = abunHouseKeeper.getDistinct(resolvedList)
print "resolvedList, len(resolvedList),len(xResolvedList), len(biResolvedCombineList) ", resolvedList, len(
resolvedList
), len(xResolvedList), len(biResolvedCombineList)
with open(folderName + "resolvedList.json", "w") as f:
json.dump(resolvedList, f)
with open(folderName + "mapDummyToRealDic.json", "w") as f:
json.dump(mapDummyToRealDic, f)
if False:
json_data = open(folderName + "resolvedList.json", "r")
resolvedList = json.load(json_data)
json_data = open(folderName + "mapDummyToRealDic.json", "r")
mapDummyToRealDic = json.load(json_data)
gapContentLookUpList = []
gapContentLookUpList = generateGapContentLookup(
folderName, mummerLink, resolvedList, contigReadGraph, contigFilename, readsetFilename, mapDummyToRealDic
)
gapContentLookUpDic = {}
gapContentLookUpList.sort()
for eachitem in gapContentLookUpList:
gapContentLookUpDic[str(eachitem[0]) + "_" + str(eachitem[1])] = [eachitem[2], eachitem[3], eachitem[4]]
print eachitem[2:4], len(eachitem[4])
with open(folderName + "gapContentLookUpDic.json", "w") as f:
json.dump(gapContentLookUpDic, f)
if False:
json_data = open(folderName + "resolvedList.json", "r")
resolvedList = json.load(json_data)
json_data = open(folderName + "mapDummyToRealDic.json", "r")
mapDummyToRealDic = json.load(json_data)
G = graphLib.seqGraph(N1 + len(mapDummyToRealDic))
addEdges(G, resolvedList)
G.condense()
G.saveToFile(folderName, "xResolvedGraph")
if False:
json_data = open(folderName + "mapDummyToRealDic.json", "r")
mapDummyToRealDic = json.load(json_data)
G = graphLib.seqGraph(0)
G.loadFromFile(folderName, "xResolvedGraph")
json_data = open(folderName + "gapContentLookUpDic.json", "r")
gapContentLookUpDic = json.load(json_data)
print "Final step: really hacking a file"
os.system("cp " + folderName + contigFilename + "_Double.fasta " + folderName + "tmpWithDummy.fasta")
contigList = IORobot.readContigsFromFile(folderName, contigFilename + "_Double.fasta")
f = open(folderName + "tmpWithDummy.fasta", "a")
for i in range(len(mapDummyToRealDic)):
id = mapDummyToRealDic[str(i)]
f.write(">SegDum" + str(i) + "\n")
f.write(contigList[id] + "\n")
f.close()
IORobot.extractGraphToContigs(
G, folderName, mummerLink, "abun.fasta", "tmpWithDummy.fasta", gapContentLookUpDic, mapDummyToRealDic
)
def abunSplitWithXResolve(folderName, mummerLink, myCountDic, contigReadGraph,
contigFilename, readsetFilename):
N1 = len(myCountDic) * 2
print "N1", N1
# Debug
G = graphLib.seqGraph(0)
G.loadFromFile(folderName, contigReadGraph)
adj = [[] for i in range(N1)]
for i in range(N1):
adj[i] = abunGraphLib.findAllReachable(i, N1, G)
Gnew = graphLib.seqGraph(N1)
for i in range(N1):
for j in adj[i]:
Gnew.insertEdge(i, j, 1)
Gnew.reportEdge()
# End Debug
if False:
json_data = open(folderName + "phaseRepeat.txt", 'r')
repeatPairs = json.load(json_data)
repeatPairs = obtainNonEmpty(repeatPairs)
biResolvedCombineList = []
for eachitem in repeatPairs:
inList, outList = eachitem[0], eachitem[1]
resolvedList = determindMatch(inList, outList, myCountDic,
folderName, contigReadGraph, N1)
biResolvedCombineList += resolvedList
### Xnode repeatResolution
xResolvedList, mapDummyToRealDic = xNodeResolving(
folderName, contigReadGraph)
### Combine resolution
resolvedList = xResolvedList + biResolvedCombineList
resolvedList = abunHouseKeeper.getDistinct(resolvedList)
print "resolvedList, len(resolvedList),len(xResolvedList), len(biResolvedCombineList) ", resolvedList, len(
resolvedList), len(xResolvedList), len(biResolvedCombineList)
with open(folderName + "resolvedList.json", 'w') as f:
json.dump(resolvedList, f)
with open(folderName + "mapDummyToRealDic.json", 'w') as f:
json.dump(mapDummyToRealDic, f)
if False:
json_data = open(folderName + "resolvedList.json", 'r')
resolvedList = json.load(json_data)
json_data = open(folderName + "mapDummyToRealDic.json", 'r')
mapDummyToRealDic = json.load(json_data)
gapContentLookUpList = []
gapContentLookUpList = generateGapContentLookup(
folderName, mummerLink, resolvedList, contigReadGraph,
contigFilename, readsetFilename, mapDummyToRealDic)
gapContentLookUpDic = {}
gapContentLookUpList.sort()
for eachitem in gapContentLookUpList:
gapContentLookUpDic[str(eachitem[0]) + "_" + str(eachitem[1])] = [
eachitem[2], eachitem[3], eachitem[4]
]
print eachitem[2:4], len(eachitem[4])
with open(folderName + "gapContentLookUpDic.json", 'w') as f:
json.dump(gapContentLookUpDic, f)
if False:
json_data = open(folderName + "resolvedList.json", 'r')
resolvedList = json.load(json_data)
json_data = open(folderName + "mapDummyToRealDic.json", 'r')
mapDummyToRealDic = json.load(json_data)
G = graphLib.seqGraph(N1 + len(mapDummyToRealDic))
addEdges(G, resolvedList)
G.condense()
G.saveToFile(folderName, "xResolvedGraph")
if False:
json_data = open(folderName + "mapDummyToRealDic.json", 'r')
mapDummyToRealDic = json.load(json_data)
G = graphLib.seqGraph(0)
G.loadFromFile(folderName, "xResolvedGraph")
json_data = open(folderName + "gapContentLookUpDic.json", 'r')
gapContentLookUpDic = json.load(json_data)
print "Final step: really hacking a file"
os.system("cp " + folderName + contigFilename + "_Double.fasta " +
folderName + "tmpWithDummy.fasta")
contigList = IORobot.readContigsFromFile(
folderName, contigFilename + "_Double.fasta")
f = open(folderName + "tmpWithDummy.fasta", 'a')
for i in range(len(mapDummyToRealDic)):
id = mapDummyToRealDic[str(i)]
f.write(">SegDum" + str(i) + "\n")
f.write(contigList[id] + "\n")
f.close()
IORobot.extractGraphToContigs(G, folderName, mummerLink, "abun.fasta",
"tmpWithDummy.fasta",
gapContentLookUpDic, mapDummyToRealDic)