def parse1KGvcf(vcffile, plines, genotypedboutput, refdboutput, altdboutput):
'''
parse1KGvcf('../1000GenomesData/CEU.low_coverage.2010_09.genotypes.vcf' , p1lines, 'testoutput', 'testoutputRef', 'testoutputAlt')
DEFAULT: only take Illumina 2.5M snps from genotype, assume they are consistant enough
'''
#>>> for l in lines:
# newg.write(l.split('\t')[0]+','+l.split('\t')[1])
afile = open('mark/arraypool/25M1.1', 'r')
lines = afile.readlines()
afile.close()
asnps = set([])
for l in lines:
asnps.add(l.split('\t')[9] + ':' + l.split('\t')[10])
print len(asnps)
vfile = open(vcffile, 'r')
vcf_reader = vcf.Reader(vfile)
poollines = gl.jsonload(plines)
outputfile = open(genotypedboutput, 'w')
ref = {}
alt = {}
for record in vcf_reader:
try:
chrom = record.INFO['GP'].split(':')[0]
pos = record.INFO['GP'].split(':')[1]
except KeyError:
continue
if chrom+':'+pos in asnps:
ref[chrom+':'+pos] = str(record.REF)
alt[chrom+':'+pos] = str(record.ALT[0])
m = chrom+':'+pos+','
sm = 0
for s in poollines:
record.genotype(s)
try:
geno = record.genotype(s)['GT']
except KeyError:
print "no genotype"
continue
if geno:
if '|' in geno:
g = geno.split('|')
if '\\' in geno:
g = geno.split('\\')
if '/' in geno:
g = geno.split('/')
an = int(g[0]) + int(g[1])
sm = sm + an
m = m + str(an) + ','
else:
m = m+str(0) + ','
if sm > 0:
outputfile.write(m.strip(',') + '\n')
gl.jsondump(ref, refdboutput)
gl.jsondump(alt, altdboutput)
def getarraysnps(report, fgenoref, fgenoalt, outname, kwargs):
'''
test this with MKReport1bysnps.txt and output of parse1KGvcf function
getarraysnps('MKReport1bysnps.txt', 'testoutputRef', 'testoutputAlt')
In this version we totally ignore compliments
Array can't have duplicates (it does). We take the last one mentioned in the array
report, because those are the 1kg ones rather than rsid
kwargs:
snp
chr
pos
theta
header
'''
print report
print kwargs
file = open(report)
lines = file.readlines()
lines.reverse()
file.close()
genoref = gl.jsonload(fgenoref)
genoalt = gl.jsonload(fgenoalt)
output = open(outname, 'w')
try:
header = kwargs['header']
h = header.split('\\t')
print h
snpi = h.index("SNP")
chri = h.index("Chr")
posi = h.index("Position")
thetai = h.index("Theta")
except KeyError:
try:
snpi = int(kwargs['snp'])
chri = int(kwargs['chr'])
posi = int(kwargs['pos'])
thetai = int(kwargs['theta'])
except KeyError:
print "No header or column numbers provided provided"
snps = set([])
freq = {}
for l in lines:
t = l.split('\t')
try:
if t[chri] not in map(lambda x: str(x), range(1,23)):
continue
else:
snppos = t[chri]+':'+t[posi]
ref = t[snpi].split('/')[0][1]
alt = t[snpi].split('/')[1][0]
if genoref[snppos] == ref and genoalt[snppos] == alt:
f = float(t[thetai])
elif genoref[snppos] == alt and genoalt[snppos] == ref:
f = 1 - float(t[thetai])
else:
continue
if f != 0 and f != 1:
#freq[snppos] = f
if snppos not in snps:
snps.add(snppos)
output.write(snppos + '\t' + str(f) + '\n')
else:
continue
except:
continue
