def main():
usage = "usage: %prog [options] arg"
parser = OptionParser(usage)
parser.add_option("--filter", type="string", dest="filter", help="extract records matching filter (default is None)", default=None)
parser.add_option("--addchr", action="store_true", dest="addchr", help="pre-pend 'chr' to chrom column ", default=False)
parser.add_option("--siteinfo", action="store_true", dest="siteinfo", help="use if vcf only has site information and lacks FORMAT column")
parser.add_option("--dump", action="store_true", dest="dump", help="dump everything after teh ID column in the 4th bed column")
parser.add_option("--chr", type="string", dest="chr", default=None, help="restrct to chromosome number specified by --chr")
(options, args)=parser.parse_args()
vcfilename=args[0]
#basename, extension = os.path.splitext(vcfilename)
#bedfile=basename+".bed"
#bedfh=open(bedfile,'w')
vcfh=open(vcfilename,'r')
#instantiate a VcfFile object
vcfobj=VcfFile(vcfilename)
#parse its metainfo lines (ones that begin with ##)
vcfobj.parseMetaAndHeaderLines(vcfh)
for dataline in vcfobj.yieldVcfDataLine(vcfh):
fields=dataline.strip().split('\t')
if options.siteinfo == True:
(chrom,pos,id,ref,alt,qual,filtercode,info)=fields[0:8]
else:
(chrom,pos,id,ref,alt,qual,filtercode,info,format)=fields[0:9]
if options.chr != None and chrom != options.chr: continue
if options.addchr ==True:
chrom='chr'+chrom
if filtercode != options.filter and options.filter != None : continue
(start,end) = (int(pos)-1, int(pos))
if options.dump == True:
# @type options
if options.siteinfo == True:
gstrings=",".join(fields[8::])
else:
gstrings=",".join(fields[9::])
dumpstring="".join([ref,alt,qual,filtercode,info,gstrings])
bedstring= "\t".join( [ chrom, str(start), str(end), id ,dumpstring] )
else:
bedstring= "\t".join( [ chrom, str(start), str(end), id] )
print bedstring
def main():
usage = "usage: %prog [options] vcf_file_one vcf|bed_file_two\n\nFind regions in the first vcf file that overlap regions of the second vcf or bed file\n"
parser = OptionParser(usage)
parser.add_option("--minCols", type="int", dest="mincols", default=1, help="mininum basepair overlap (default is one)")
parser.add_option("--v", action="store_true", dest="reverse", help="Print regions in first vcf that DO NOT overlap second vcf|bed file")
parser.add_option("--filter", type="string", dest="filter", default=None, help="intersect records only set with filter (default is None")
parser.add_option("--info", type="string", dest="infotag", help="INFO tag id that annotates what type of variant the VCF record is", default="TYPE")
parser.add_option("--type", type="string", dest="variantype", help="type of variant (SNP INS DEL)", default=None)
parser.add_option("--noheader", action="store_true", dest="noheader", help="VCF file one has no header line", default=False)
parser.add_option("--nochrprefix", action="store_false", dest="chrprefix", help="use if the bed doesn't have chr prefix in chrom column", default=True)
(options, args)=parser.parse_args()
sys.stderr.write("intersecting two files ...\n")
vcf_file_one=args[0]
in2_fname=args[1]
in2_fname_ext= os.path.splitext(in2_fname)[1][1:]
if "bed" == in2_fname_ext:
bitsets = binned_bitsets_from_file( open( in2_fname ) )
if "vcf" == in2_fname_ext:
bitsets = binned_bitsets_from_vcffile( in2_fname , options.filter)
vcfobj=VcfFile(vcf_file_one)
vcfh=open(vcf_file_one,'r')
if options.noheader == False:
vcfobj.parseMetaAndHeaderLines(vcfh)
header=vcfobj.returnHeader()
#print header
#vcfobj.parseMetaAndHeaderLines(vcfh)
#descriptors = vcfobj.getMetaInfoDescription()
#infoids=[]
#for (tag, description) in descriptors:
# infoids.append(tag)
#if options.infotag not in infoids and options.infotag != 'QUAL' and options.infotag != "" and options.noheader == False:
# sys.stderr.write(options.infotag + " tag not in ##INFO headers!\n")
# exit(1)
print header
for dataline in vcfobj.yieldVcfDataLine(vcfh):
fields=dataline.strip().split('\t')
(chrom,pos,id,ref,alt,qual,filtercode,info)=fields[0:8]
(start,end) = (int(pos)-1, int(pos))
#pass the filter code
if filtercode != options.filter and options.filter != None:
continue
#check to see if record is the correct variant TYPE
if options.variantype != None:
pattern=options.infotag+'=('+options.variantype+')'
if re.search(pattern, info ) == None:
continue
if options.chrprefix == True:
chrom="chr"+chrom
if chrom in bitsets and bitsets[chrom].count_range( start, end-start ) >= options.mincols:
if not options.reverse:
print dataline
else:
if options.reverse == True:
print dataline
def main():
usage = "usage: %prog [options] maf file.vcf"
parser = OptionParser(usage)
parser.add_option("--maftag", type="string", dest="maftag", help="INFO tag id that annotates the allele freq of the record", default="AF")
parser.add_option("--variantag", type="string", dest="vtag", help="INFO tag that annotates the type of variant type", default="VT")
parser.add_option("--variantype", type="string", dest="variantype", help="type of variant (SNP INS DEL)", default=None)
parser.add_option("--filter", type="string", dest="filter", help="extract records matching filter (default is None)", default=None)
parser.add_option("--noheader", action="store_true", dest="noheader", help="VCF file has no header file", default=False)
parser.add_option("--quiet", action="store_true", dest="quiet", help="don't print vcf output to stdout", default=False)
parser.add_option("--leq", type="float", dest="leq", default=1.0, help="keep variants with AF <= (default 1)")
parser.add_option("--geq", type="float", dest="geq", default=0.0, help="keep variants with AF >= (default 0)")
(options, args)=parser.parse_args()
if len(args)!=1:
sys.stderr.write(usage+"\n")
exit(1)
vcfilename=args[0]
#maf=float(args[0])
freqfh=open('freq.log', 'w')
vcfh=open(vcfilename,'r')
#instantiate a VcfFile object
vcfobj=VcfFile(vcfilename)
#parse its metainfo lines (ones that begin with ##)
if options.noheader == False:
vcfobj.parseMetaLines(vcfh)
#vcfobj.printMetaLines()
descriptors = vcfobj.getMetaInfoDescription()
infoids=[]
for (tag, description) in descriptors:
infoids.append(tag)
if options.maftag not in infoids and options.maftag != 'QUAL' and options.noheader == False:
sys.stderr.write(options.maftag + " tag not in ##INFO headers!\n")
exit(1)
if options.vtag not in infoids and options.vtag != 'QUAL' and options.noheader==False:
sys.stderr.write(options.vtag + " tag not in ##INFO headers!\n")
exit(1)
#vcfh.seek(0)
if options.noheader == False:
vcfobj.parseHeaderLine(vcfh)
if options.variantype==None:
variantpattern=options.vtag+'=(\w+);'
else:
variantpattern=options.vtag+'=('+options.variantype+');'
mafpattern=options.maftag+'=(0.\d+)'
#print mafpattern, variantpattern
for dataline in vcfobj.yieldVcfDataLine(vcfh):
#print dataline
fields=dataline.strip().split('\t')
(chrom,pos,id,ref,alt,qual,filtercode,info)=fields[0:8]
#if filtercode != options.filter and options.filter != None : continue
if re.search(variantpattern, info ) == None:
#sys.stderr.write("no variant pattern\n")
continue
variant_type=re.search(variantpattern, info ).groups()[0]
if re.search(mafpattern, info ) == None:
#sys.stderr.write("No mafpattern!\n")
#sys.stderr.write(dataline+"\n")
continue
maf_value=re.search(mafpattern, info ).groups()[0]
if float(maf_value) <= options.leq and float(maf_value) >= options.geq:
if options.quiet == False:
print dataline
logstring="\t".join([chrom,pos,id,ref,alt,variant_type, options.maftag, maf_value])
freqfh.write(logstring+'\n')
def main():
usage = "usage: %prog [options] file.vcf\n print records belonging to a certain type of variant class (e.g. SNP) in a VCF file\n\n"
parser = OptionParser(usage)
parser.add_option(
"--info",
type="string",
dest="infotag",
help="INFO tag id that annotates what type of variant the VCF record is",
default="TYPE",
)
parser.add_option("--type", type="string", dest="variantype", help="type of variant (SNP INS DEL)", default=None)
parser.add_option(
"--filter", type="string", dest="filter", help="extract records matching filter (default is None)", default=None
)
parser.add_option("--noheader", action="store_true", dest="noheader", help="VCF file has no header file")
(options, args) = parser.parse_args()
if options.infotag == "":
sys.stderr.write("provide a value for --info parameter!\n")
exit(1)
if options.variantype == "":
sys.stderr.write("provide a value of --type parameter!\n")
exit(1)
variant_dict = {}
vcfilename = args[0]
vcfh = open(vcfilename, "r")
# instantiate a VcfFile object
vcfobj = VcfFile(vcfilename)
# parse its metainfo lines (ones that begin with ##)
vcfobj.parseMetaAndHeaderLines(vcfh)
vcfobj.printMetaAndHeaderLines()
descriptors = vcfobj.getMetaInfoDescription()
infoids = []
for (tag, description) in descriptors:
infoids.append(tag)
if options.infotag not in infoids and options.infotag != "QUAL":
sys.stderr.write(options.infotag + " tag not in ##INFO headers!\n")
exit(1)
if options.variantype != None:
pattern = options.infotag + "=(" + options.variantype + ")"
for dataline in vcfobj.yieldVcfDataLine(vcfh):
fields = dataline.strip().split("\t")
(chrom, pos, id, ref, alt, qual, filtercode, info) = fields[0:8]
if filtercode != options.filter and options.filter != None:
continue
if options.variantype != None:
if re.search(pattern, info) == None:
continue
else:
value = re.search(pattern, info).groups()[0]
print dataline
else:
print dataline
