def convertAlignmentReadGroup2UCLAIDInVCF(self, inputFname, outputFname, minDepth=1, includeIndels=False,\
maxContigNumber=None):
"""
2012.5.10
"""
sys.stderr.write("Converting %s from VCF to EigenStrat ...\n"%(inputFname))
vcfFile = VCFFile(inputFname=inputFname, minDepth=minDepth)
#replace Variant/PooledTissues/2002053/genome.algn.split.part17/5tissues.pooled.rmdup.bam with just monkey ID
newSampleIDHeader = []
for sampleID in vcfFile.sampleIDHeader:
readGroupData = VervetDB.VervetDB.parseAlignmentReadGroupWithoutDB(sampleID)
UCLAID = readGroupData.individual_code
newSampleIDHeader.append(UCLAID)
#new header for every output contig
newHeader = vcfFile.header[:vcfFile.sampleStartingColumn] + newSampleIDHeader
counter = 0
real_counter = 0
outVCFFile = VCFFile(outputFname=outputFname)
outVCFFile.metaInfoLs = vcfFile.metaInfoLs
outVCFFile.header = newHeader
outVCFFile.writeMetaAndHeader()
for vcfRecord in vcfFile.parseIter():
counter += 1
if not includeIndels and (len(vcfRecord.refBase)!=1 or len(vcfRecord.altBase)!=1):
#it's an indel if refBase or altBase is not just one base
continue
chr = vcfRecord.chr
if maxContigNumber:
contigNumber = int(self.contig_number_pattern.search(chr).group('contigNumber'))
if contigNumber>maxContigNumber:
continue
real_counter += 1
# set genotype whose depth is below minDepth to ./. (=missing)
for i in xrange(1, len(vcfRecord.data_row)): #[0] is the ref base
callData = vcfRecord.data_row[i]
if callData is None or callData.get('DP',0)<minDepth:
sampleColumnIndex = i+vcfFile.sampleStartingColumn-1
vcfRecord.row[sampleColumnIndex] = './.'
outVCFFile.writeVCFRecord(vcfRecord)
vcfFile.close()
#close all output files
outVCFFile.close()
sys.stderr.write("%s (out of %s) loci.\n"%(real_counter, counter))
def splitNamVCFIntoMultipleSingleChrVCF(self, inputFname, outputDir, minDepth=1, includeIndels=False, maxContigNumber=1000):
"""
2012.5.10
Two things in Nam's VCF file are to be modified.
1. extract VRC UCLAID from its sample ID
2. replace vervet1_scaffolds_Contig137 with simply "Contig137"
"""
sys.stderr.write("Converting %s from VCF to EigenStrat ...\n"%(inputFname))
from pymodule.VCFFile import VCFFile
vcfFile = VCFFile(inputFname=inputFname, minDepth=minDepth)
#replace Variant/PooledTissues/2002053/genome.algn.split.part17/5tissues.pooled.rmdup.bam with just monkey ID
import re
newSampleIDHeader = []
for sampleID in vcfFile.sampleIDHeader:
search_result = self.UCLAID_Pattern.search(sampleID)
UCLAID = search_result.group('UCLAID')
newSampleIDHeader.append(UCLAID)
#new header for every output contig
newHeader = vcfFile.header[:vcfFile.sampleStartingColumn] + newSampleIDHeader
chr2outVCFFile = {}
counter = 0
real_counter = 0
for vcfRecord in vcfFile.parseIter():
counter += 1
if not includeIndels and (len(vcfRecord.refBase)!=1 or len(vcfRecord.altBase)!=1):
#it's an indel if refBase or altBase is not just one base
continue
contig_id_pattern_result = self.contig_id_pattern.search(vcfRecord.chr)
chr = contig_id_pattern_result.group('contigID')
if maxContigNumber:
contigNumber = int(self.contig_number_pattern.search(chr).group('contigNumber'))
if contigNumber>maxContigNumber:
continue
real_counter += 1
vcfRecord.chr = chr
pos = vcfRecord.pos
if chr not in chr2outVCFFile:
outputFname = os.path.join(outputDir, '%s.vcf'%(chr))
outVCFFile = VCFFile(outputFname=outputFname)
outVCFFile.metaInfoLs = vcfFile.metaInfoLs
outVCFFile.header = newHeader
outVCFFile.writeMetaAndHeader()
chr2outVCFFile[chr] = outVCFFile
outVCFFile = chr2outVCFFile.get(chr)
# set genotype whose depth is below minDepth to ./. (=missing)
for i in xrange(1, len(vcfRecord.data_row)): #[0] is the ref base
callData = vcfRecord.data_row[i]
if callData is None or callData.get('DP',0)<minDepth:
sampleColumnIndex = i+vcfFile.sampleStartingColumn-1
vcfRecord.row[sampleColumnIndex] = './.'
outVCFFile.writeVCFRecord(vcfRecord)
vcfFile.close()
#close all output files
for chr, outVCFFile in chr2outVCFFile.iteritems():
outVCFFile.close()
sys.stderr.write("%s (out of %s) loci from %s chromosomes.\n"%(real_counter, counter, len(chr2outVCFFile)))