def run(self):
"""
"""
if self.debug:
import pdb
pdb.set_trace()
reader = MatrixFile(self.inputFname)
reader.constructColName2IndexFromHeader()
noOfMendelErrorColumnIndex = reader.getColIndexGivenColHeader(colHeader='N')
SNPIDColumnIndex = reader.getColIndexGivenColHeader(colHeader='SNP')
writer = csv.writer(open(self.outputFname, 'w'), delimiter='\t')
header = ['chromosome', 'position', 'noOfMendelErrors']
writer.writerow(header)
counter = 0
real_counter = 0
for row in reader:
SNPID = row[SNPIDColumnIndex]
noOfMendelErrors = int(row[noOfMendelErrorColumnIndex])
if noOfMendelErrors <=self.maxNoOfMendelError:
chr, pos = SNPID.split('_')
data_row = [chr, pos, noOfMendelErrors]
writer.writerow(data_row)
real_counter += 1
counter += 1
del reader
del writer
sys.stderr.write("%s/%s lines outputted.\n"%(real_counter, counter))
def getLocusID2MissingFraction(self, inputFname=None):
"""
2014.01.08
"""
sys.stderr.write("Reading in the missing statistics from %s ... " %
(inputFname))
locusID2Stat = {}
reader = MatrixFile(inputFname=inputFname)
reader.constructColName2IndexFromHeader()
locusIDIndex = reader.getColIndexGivenColHeader("locusID")
statIndex = reader.getColIndexGivenColHeader("occurrence_byFixedValue")
counter = 0
for row in reader:
locusID = row[locusIDIndex]
chromosome, start = locusID.split('_')[:2]
start = int(start)
stat = float(row[statIndex])
key = (chromosome, start, start)
if key in locusID2Stat:
if stat < locusID2Stat[key]:
#take lowest value
locusID2Stat[key] = stat
else:
locusID2Stat[key] = stat
counter += 1
del reader
sys.stderr.write(
" %s unique loci with missing fraction out of %s total loci.\n" %
(len(locusID2Stat), counter))
return locusID2Stat
def getSampleID2IndividualData_UNGC(self, inputFname=None):
"""
2013.04.04
Format is like this from UNGC = UCLA Neuroscience Genomics Core:
FCID Lane sample ID sample code sample name Index Description SampleProject
D1HYNACXX 1 Ilmn Human control pool ( 4plex) IP1 INDEX IS UNKNOWN prepared by Illumina (4 plex pool) 2013-029A
D1HYNACXX 2 UNGC Human Sample 1 S1 AS001A ATTACTCG TruSeq DNA PCR Free beta kit 2013-029A
"""
sys.stderr.write("Getting sampleID2IndividualData from %s ..."%(inputFname))
sampleID2IndividualData = {}
reader = MatrixFile(inputFname, openMode='r', delimiter=',')
reader.constructColName2IndexFromHeader()
sampleIDIndex = reader.getColIndexGivenColHeader("sample ID")
sampleNameIndex = reader.getColIndexGivenColHeader("sample name")
libraryIndexIndex = reader.getColIndexGivenColHeader("Index")
for row in reader:
sampleID = row[sampleIDIndex].replace(' ', '_') #2013.04.04 stupid quirks
sampleName = row[sampleNameIndex]
libraryIndex = row[libraryIndexIndex]
if sampleID not in sampleID2IndividualData:
sampleID2IndividualData[sampleID] = PassingData(sampleName=sampleName, libraryIndexList=[])
if sampleName!=sampleID2IndividualData[sampleID].sampleName:
sys.stderr.write("Error: sampleID %s is associated with two different sample names (%s, %s).\n"%\
(sampleID, sampleName, sampleID2IndividualData[sampleID].sampleName))
raise
sampleID2IndividualData[sampleID].libraryIndexList.append(libraryIndex)
sys.stderr.write("%s entries.\n"%(len(sampleID2IndividualData)))
return sampleID2IndividualData
def constructPedigreeGraphFromPOEdgeFile(self, inputFname=None):
"""
2012.8.23
inputFname is output of vervet/src/pedigree/DiscoverParentOffspringFromPlinkIBD.py
"""
sys.stderr.write("Constructing pedigree-graph out of %s ..."%(inputFname))
DG=nx.DiGraph()
reader = None
childNodeSet = set()
reader = MatrixFile(inputFname)
reader.constructColName2IndexFromHeader()
parentIDIndex = reader.getColIndexGivenColHeader("parentID")
childIDIndex = reader.getColIndexGivenColHeader("childID")
distToPOVectorIndex = reader.getColIndexGivenColHeader("distToPOVector")
counter = 0
for row in reader:
childID = row[childIDIndex]
childNodeSet.add(childID)
parentID = row[parentIDIndex]
distToPOVector = float(row[distToPOVectorIndex])
DG.add_edge(parentID, childID, weight=distToPOVector)
counter += 1
del reader
sys.stderr.write("%s children, %s nodes. %s edges. %s connected components.\n"%(\
len(childNodeSet), DG.number_of_nodes(), DG.number_of_edges(), \
nx.number_connected_components(DG.to_undirected())))
return PassingData(DG=DG, childNodeSet=childNodeSet)
def getMendelErrorIndividualLocusData(self, mendelErrorFname=None, individualID2Index=None):
"""
2013.1.29
"""
sys.stderr.write("Getting data on loci involved in mendel-errors from %s ..."%(mendelErrorFname))
locus_id2individual_index_ls = {}
#inf = utils.openGzipFile(mendelErrorFname, 'r')
reader = MatrixFile(inputFname=mendelErrorFname)
#header = reader.next()
reader.constructColName2IndexFromHeader()
counter = 0
for row in reader:
individual_id = row[reader.getColIndexGivenColHeader('KID')]
if individual_id in individualID2Index:
index =individualID2Index.get(individual_id)
else:
sys.stderr.write("Individual %s not in individualID2Index.\n"%(individual_id))
sys.exit(3)
snp_id = row[3]
if snp_id not in locus_id2individual_index_ls:
locus_id2individual_index_ls[snp_id] = []
locus_id2individual_index_ls[snp_id].append(index)
counter += 1
del reader
sys.stderr.write(" %s calls of %s loci, involved in mendel errors.\n"%\
(counter, len(locus_id2individual_index_ls)))
return locus_id2individual_index_ls
def getMonkeyID2Coverage(self, inputFname):
"""
2012.9.4
copied from vervet/src/misc.py
2012.2.10
inputFname is output of SequencingStrategy.assignVRCSequencePriorityBasedOnPedigree() + manual change of top ones
"""
sys.stderr.write("Reading the list of ranked monkeys from %s ..." % (inputFname))
reader = MatrixFile(inputFname)
reader.constructColName2IndexFromHeader()
monkey_id_index = reader.getColIndexGivenColHeader("UCLAID")
pre_set_coverage_index = reader.getColIndexGivenColHeader("pre-set-coverage")
future_coverage_index = reader.getColIndexGivenColHeader("future coverage")
to_sequence_monkey_id2coverage = {}
for row in reader:
monkey_id = row[monkey_id_index]
pre_set_coverage = row[pre_set_coverage_index]
if pre_set_coverage:
pre_set_coverage = float(pre_set_coverage)
else:
pre_set_coverage = 0
future_coverage = 0
if len(row) >= future_coverage_index + 1:
future_coverage = float(row[future_coverage_index])
to_sequence_monkey_id2coverage[monkey_id] = max(future_coverage, pre_set_coverage)
del reader
sys.stderr.write(" %s monkeys are to-be-sequenced.\n" % (len(to_sequence_monkey_id2coverage)))
return to_sequence_monkey_id2coverage
"""
def run(self):
"""
2013.07.24
"""
if self.debug:
import pdb
pdb.set_trace()
#inf = utils.openGzipFile(self.inputFname)
reader = MatrixFile(inputFname=self.inputFname)
reader.constructColName2IndexFromHeader()
writer = MatrixFile(inputFname=self.outputFname,
openMode='w',
delimiter='\t')
header = ["SNPID", "oldChromosome", "Chromosome", "Start", "Stop", "N"]
writer.writeHeader(header)
counter = 0
for row in reader:
new_row = self.processRow(row)
writer.writerow(new_row)
counter += 1
sys.stderr.write("%s lines processed.\n" % (counter))
del reader
del writer
def constructPedigreeGraphFromPlinkIBD(self, inputFname=None, maxDistanceToPOVector=0.04, drawDistribution=False, outputFnamePrefix=None):
"""
2012.8.14
"""
sys.stderr.write("Constructing pedigree-graph out of plink-ibd %s ..."%(inputFname))
DG=nx.DiGraph()
childNodeSet = set()
reader = MatrixFile(inputFname)
reader.constructColName2IndexFromHeader()
monkey1IDIndex = reader.getColIndexGivenColHeader("IID1")
monkey2IDIndex = reader.getColIndexGivenColHeader("IID2")
Z0Index = reader.getColIndexGivenColHeader("Z0")
Z1Index = reader.getColIndexGivenColHeader("Z1")
Z2Index = reader.getColIndexGivenColHeader("Z2")
poVector = numpy.array([0,1,0.0])
counter = 0
real_counter = 0
data_ls = []
for row in reader:
monkey1ID = int(row[monkey1IDIndex]) #turn it into integer so could compare age
monkey2ID = int(row[monkey2IDIndex])
Z0 = float(row[Z0Index])
Z1 = float(row[Z1Index])
Z2 = float(row[Z2Index])
ZVector = numpy.array([Z0, Z1, Z2])
dist = numpy.linalg.norm(poVector-ZVector)
if drawDistribution and outputFnamePrefix:
data_ls.append(dist)
if dist<=maxDistanceToPOVector:
if monkey1ID>monkey2ID:
childID = monkey1ID
parentID = monkey2ID
else:
childID = monkey2ID
parentID = monkey1ID
DG.add_edge(parentID, childID, weight=dist)
childNodeSet.add(childID)
real_counter += 1
counter += 1
del reader
sys.stderr.write("%s out of %s lines become PO pairs. %s children, %s nodes. %s edges. %s connected components.\n"%(\
real_counter, counter, len(childNodeSet), DG.number_of_nodes(), DG.number_of_edges(), \
nx.number_connected_components(DG.to_undirected())))
if drawDistribution and outputFnamePrefix:
outputFname = '%s_IBDVector2POVectorDist_hist.png'%(outputFnamePrefix)
yh_matplotlib.drawHist(data_ls, title='', \
xlabel_1D="dist(ZVector,POVector)", xticks=None, \
outputFname=outputFname, min_no_of_data_points=10, \
needLog=True, \
dpi=200, min_no_of_bins=25)
return PassingData(DG=DG, childNodeSet=childNodeSet)
def readInCoordinateMap(self, coordinateMapFname=None):
"""
2013.07.11
querySNPID queryStrand queryChromosome queryStart queryStop queryRefBase queryAltBase queryAlignmentSpan
queryAlignmentStart queryAlignmentStop newChr newRefStart newRefStop newRefBase targetAlignmentSpan
targetAlignmentStart targetAlignmentStop
"""
sys.stderr.write("Reading in the coordinate map from %s ..." %
(coordinateMapFname))
oldCoordinate2newCoordinateDataLs = {}
reader = MatrixFile(inputFname=coordinateMapFname)
reader.constructColName2IndexFromHeader()
oldChromosomeIndex = reader.getColIndexGivenColHeader(
"queryChromosome")
oldStartIndex = reader.getColIndexGivenColHeader("queryStart")
strandIndex = reader.getColIndexGivenColHeader("queryStrand")
oldRefBaseIndex = reader.getColIndexGivenColHeader("queryRefBase")
oldAltBaseIndex = reader.getColIndexGivenColHeader("queryAltBase")
newChromosomeIndex = reader.getColIndexGivenColHeader("newChr")
newStartIndex = reader.getColIndexGivenColHeader("newRefStart")
newStopIndex = reader.getColIndexGivenColHeader("newRefStop")
newRefBaseIndex = reader.getColIndexGivenColHeader("newRefBase")
counter = 0
for row in reader:
oldChromosome = row[oldChromosomeIndex]
oldStart = int(row[oldStartIndex])
strand = row[strandIndex]
oldRefBase = row[oldRefBaseIndex]
oldAltBase = row[oldAltBaseIndex]
newChromosome = row[newChromosomeIndex]
newStart = int(row[newStartIndex])
newStop = int(row[newStopIndex])
newRefBase = row[newRefBaseIndex]
key = (oldChromosome, oldStart)
if key not in oldCoordinate2newCoordinateDataLs:
oldCoordinate2newCoordinateDataLs[key] = []
oldCoordinate2newCoordinateDataLs[key].append(PassingData(strand=strand, oldRefBase=oldRefBase, \
oldAltBase=oldAltBase, newChromosome=newChromosome, newStart=newStart,\
newStop=newStop, newRefBase=newRefBase))
counter += 1
del reader
sys.stderr.write("%s old coordinates with %s new coordinates.\n" %
(len(oldCoordinate2newCoordinateDataLs), counter))
return oldCoordinate2newCoordinateDataLs
def getMonkeyPair2IBDVector(self, inputFname=None):
"""
2012.9.10
return monkeyIDSet as well
2012.9.6
"""
sys.stderr.write("Getting monkey pair 2 IBD vector from %s ..." % (inputFname))
reader = MatrixFile(inputFname)
reader.constructColName2IndexFromHeader()
monkey1IDIndex = reader.getColIndexGivenColHeader("IID1")
monkey2IDIndex = reader.getColIndexGivenColHeader("IID2")
IBDIndex = reader.getColIndexGivenColHeader("PI_HAT")
Z0Index = reader.getColIndexGivenColHeader("Z0")
Z1Index = reader.getColIndexGivenColHeader("Z1")
Z2Index = reader.getColIndexGivenColHeader("Z2")
formatFunc = lambda x: "%.2f" % (x)
monkeyPair2IBDVector = {}
counter = 0
monkeyIDSet = set()
for row in reader:
monkey1ID = row[monkey1IDIndex]
monkey2ID = row[monkey2IDIndex]
monkey_id_pair = [monkey1ID, monkey2ID]
monkey_id_pair.sort()
key = tuple(monkey_id_pair)
Z0 = float(row[Z0Index])
Z1 = float(row[Z1Index])
Z2 = float(row[Z2Index])
IBD = float(row[IBDIndex])
IBDVector = [Z0, Z1, Z2]
IBDVector = map(formatFunc, IBDVector)
IBDVectorStr = ",".join(IBDVector)
data = PassingData(IBD=IBD, IBDVector=IBDVector, IBDVectorStr=IBDVectorStr)
if key in monkeyPair2IBDVector:
sys.stderr.write(
"WARNING: key %s has value %s in monkeyPair2IBDVector already. value overwritten with %s.\n"
% (repr(key), monkeyPair2IBDVector.get(key), data)
)
monkeyPair2IBDVector[key] = data
monkeyIDSet.add(monkey1ID)
monkeyIDSet.add(monkey2ID)
counter += 1
sys.stderr.write(
" %s pairs of IBD vectors for %s unique monkeys.\n" % (len(monkeyPair2IBDVector), len(monkeyIDSet))
)
return PassingData(monkeyPair2IBDVector=monkeyPair2IBDVector, monkeyIDSet=monkeyIDSet)
def setup(self, **keywords):
"""
"""
AbstractMatrixFileWalker.setup(self, **keywords)
#construct a individualCode2readGroup from readGroupFname
self.invariantPData.individualCode2readGroup = {}
reader = MatrixFile(inputFname=self.readGroupFname)
reader.constructColName2IndexFromHeader()
if self.readGroupHeader:
readGroupIndex = reader.getColIndexGivenColHeader(self.readGroupHeader)
else:
readGroupIndex = 0
for row in reader:
readGroup = row[readGroupIndex]
individualAlignment = self.db_vervet.parseAlignmentReadGroup(readGroup).individualAlignment
if individualAlignment:
individual_code = individualAlignment.individual_sequence.individual.code
self.invariantPData.individualCode2readGroup[individual_code] = readGroup
del reader
return 1
def getLocusNewID2mapPvalue(self, liftOverLocusMapPvalueFname=None):
"""
2014.01.04
oldChromosome, oldStart, oldStop, oldStrand, newChromosome, newStart, newStop, mapPvalue
"""
sys.stderr.write("Reading in the coordinate map from %s ..." %
(liftOverLocusMapPvalueFname))
locusNewID2mapPvalue = {}
reader = MatrixFile(inputFname=liftOverLocusMapPvalueFname)
reader.constructColName2IndexFromHeader()
strandIndex = reader.getColIndexGivenColHeader("oldStrand")
newChromosomeIndex = reader.getColIndexGivenColHeader("newChromosome")
newStartIndex = reader.getColIndexGivenColHeader("newStart")
newStopIndex = reader.getColIndexGivenColHeader("newStop")
mapPvalueIndex = reader.getColIndexGivenColHeader("mapPvalue")
counter = 0
for row in reader:
strand = row[strandIndex]
newChromosome = row[newChromosomeIndex]
newStart = int(row[newStartIndex])
newStop = int(row[newStopIndex])
mapPvalue = float(row[mapPvalueIndex])
key = (newChromosome, newStart, newStop)
if key in locusNewID2mapPvalue:
if mapPvalue < locusNewID2mapPvalue[key]:
#take lowest value
locusNewID2mapPvalue[key] = mapPvalue
else:
locusNewID2mapPvalue[key] = mapPvalue
counter += 1
del reader
sys.stderr.write(
"%s unique loci with map p-value out of %s total loci.\n" %
(len(locusNewID2mapPvalue), counter))
return locusNewID2mapPvalue
def readInStats(self, inputFname=None):
"""
2013.07.15
"""
sys.stderr.write("Reading stats from %s ..." % (inputFname))
data_matrix = []
reader = MatrixFile(inputFname)
reader.constructColName2IndexFromHeader()
switchFrequencyIndex = reader.getColIndexGivenColHeader(
"noOfSwitchPoints_by_noOfLociWithUniqueHit")
regionSpanIndex = reader.getColIndexGivenColHeader("regionSpan")
noOfLociIndex = reader.getColIndexGivenColHeader("#sitesInInput2")
totalSpan = 0
totalNoOfLoci = 0
counter = 0
for row in reader:
counter += 1
switchFrequency = row[switchFrequencyIndex]
regionSpan = row[regionSpanIndex]
noOfLoci = row[noOfLociIndex]
if switchFrequency and regionSpan and noOfLoci: #non-empty
switchFrequency = float(switchFrequency)
regionSpan = int(float(regionSpan))
noOfLoci = int(float(noOfLoci))
data_matrix.append([switchFrequency, regionSpan, noOfLoci])
totalSpan += regionSpan
totalNoOfLoci += noOfLoci
reader.close()
sys.stderr.write(" %s valid entries (from %s rows) with totalSpan=%s, totalNoOfLoci=%s.\n"%\
(len(data_matrix), counter, totalSpan, totalNoOfLoci))
return PassingData(data_matrix=data_matrix,
totalSpan=totalSpan,
totalNoOfLoci=totalNoOfLoci)
