def getLocusID2MissingFraction(self, inputFname=None):
"""
2014.01.08
"""
sys.stderr.write("Reading in the missing statistics from %s ... "%(inputFname))
locusID2Stat = {}
reader = MatrixFile(path=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 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(path=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)
def readInSwitchDensity(self, inputFname=None):
"""
2013.07.11
"""
sys.stderr.write("Reading in switch density from %s ..." %
(inputFname))
reader = MatrixFile(path=inputFname)
reader.constructColName2IndexFromHeader()
noOfSwitchesPerLocusIndex = reader.getColIndexGivenColHeader(
"noOfSwitchesPerLocus")
counter = 0
real_counter = 0
switchDensity = 0
for row in reader:
switchDensity = float(row[noOfSwitchesPerLocusIndex])
counter += 1
break
del reader
return PassingData(switchDensity=switchDensity)
def run(self):
"""
"""
if self.debug:
import pdb
pdb.set_trace()
reader = MatrixFile(path=self.inputFname)
reader.constructColName2IndexFromHeader()
meanMendelErrorIndex = reader.getColIndexGivenColHeader(
"meanMendelError")
noOfLociIndex = reader.getColIndexGivenColHeader("sampled_base_count")
sumOfMendelErrorIndex = reader.getColIndexGivenColHeader(
"sumOfMendelError")
plinkPedigreeFile = PlinkPedigreeFile(path=self.pedigreeFname)
familyStructureData = plinkPedigreeFile.getFamilyStructurePlinkWay()
twoParentFamilyCountData = self.getNoOfFamiliesAndKidsGivenParentSetSize(noOfParents2FamilyData=familyStructureData.noOfParents2FamilyData, \
parentSetSize=2)
singleParentFamilyCountData = self.getNoOfFamiliesAndKidsGivenParentSetSize(noOfParents2FamilyData=familyStructureData.noOfParents2FamilyData, \
parentSetSize=1)
zeroParentFamilyCountData = self.getNoOfFamiliesAndKidsGivenParentSetSize(noOfParents2FamilyData=familyStructureData.noOfParents2FamilyData, \
parentSetSize=0)
writer = MatrixFile(self.outputFname, mode='w', delimiter='\t')
header = ["ID", "noOfTotalLoci", \
"noOfTwoParentFamilies", "noOfParentsInTwoParentFamilies", "noOfKidsInTwoParentFamilies", "noOfIndividualsInTwoParentFamilies", \
"noOfSingleParentFamilies", "noOfParentsInSingleParentFamilies", "noOfKidsInSingleParentFamilies", "noOfIndividualsInSingleParentFamilies", \
"noOfZeroParentFamilies", "noOfParentsInZeroParentFamilies", "noOfKidsInZeroParentFamilies", "noOfIndividualsInZeroParentFamilies", \
"noOfTotalMendelErrors", \
"noOfMendelErrorsPerLocusPerNuclearFamily", "noOfMendelErrorsPerNuclearFamily"]
writer.writeHeader(header)
for row in reader:
meanMendelError = float(row[meanMendelErrorIndex])
noOfLoci = int(row[noOfLociIndex])
sumOfMendelError = int(row[sumOfMendelErrorIndex])
noOfNuclearFamilies = twoParentFamilyCountData.noOfFamilies
if noOfNuclearFamilies > 0:
noOfMendelErrorsPerLocusPerNuclearFamily = meanMendelError / float(
noOfNuclearFamilies)
noOfMendelErrorsPerNuclearFamily = sumOfMendelError / float(
noOfNuclearFamilies)
else:
noOfMendelErrorsPerLocusPerNuclearFamily = -1
noOfMendelErrorsPerNuclearFamily = -1
data_row = [row[0], noOfLoci, \
noOfNuclearFamilies, twoParentFamilyCountData.noOfParents, twoParentFamilyCountData.noOfKids, \
twoParentFamilyCountData.noOfIndividuals,\
singleParentFamilyCountData.noOfFamilies, singleParentFamilyCountData.noOfParents, singleParentFamilyCountData.noOfKids,\
singleParentFamilyCountData.noOfIndividuals,\
zeroParentFamilyCountData.noOfFamilies, zeroParentFamilyCountData.noOfParents, zeroParentFamilyCountData.noOfKids,\
zeroParentFamilyCountData.noOfIndividuals,\
sumOfMendelError, \
noOfMendelErrorsPerLocusPerNuclearFamily,noOfMendelErrorsPerNuclearFamily ]
writer.writerow(data_row)
plinkPedigreeFile.close()
reader.close()
writer.close()