def _secondRun_(srOutput,
srInput,
srTopQuantile,
srWindowSize,
snpsds,
phed,
p_i,
binary=False):
"""
Sets up a second run, using and/or on SNPs.
"""
print "Preparing a second stage analyzis."
sys.stdout.flush()
#Collect top n% SNPs.
print "Loading first stage results."
sys.stdout.flush()
result = gwaResults.SNPResult(resultFile=srInput,
snpsds=snpsds,
phenotypeID=p_i)
# - filter results
result.negLogTransform()
res = SecondStageAnalysis.retrieveSecondRunSNPs(result, srTopQuantile,
srWindowSize)
snps = res['snps']
snpTypes = res['snpTypes']
chromosomes = res['chromosomes']
positions = res['positions']
marfs = res['marfs']
mafs = res['mafs']
if binary:
#Run Fisher's Exact Test:
sys.stdout.write("Running Fisher's exact test (in R).\n")
sys.stdout.flush()
phenVals = phed.getPhenVals(p_i)
pvals = run_fet(snps, phenVals)
#print zip(positions,pvals,genotype_var_perc)
else:
#Run Emma:
sys.stdout.write("Running Kruskal-Wallis (in R).\n")
sys.stdout.flush()
phenVals = phed.getPhenVals(p_i)
pvals = run_kw(snps, phenVals)
#print zip(positions,pvals,genotype_var_perc)
sys.stdout.write("Writing results to file.\n")
sys.stdout.flush()
#Write results to file!
f = open(srOutput, "w")
f.write("Chromosome,position,p-value,marf,maf,snpType,second_pos\n")
for (chr, pos, pval, marf, maf, snpType) in zip(chromosomes, positions,
pvals, marfs, mafs,
snpTypes):
f.write(
str(chr) + "," + str(pos[0]) + "," + str(pval) + "," + str(marf) +
"," + str(maf) + "," + str(snpType) + "," + str(pos[1]) + "\n")
f.close()
def _secondRun_(srOutput, srInput, srTopQuantile, srWindowSize, snpsds, phed,
p_i, kinshipSnpsDataset):
"""
Sets up a second run, using and/or on SNPs.
"""
print "Preparing a second stage analyzis."
sys.stdout.flush()
#Collect top n% SNPs.
print "Loading first stage results."
sys.stdout.flush()
result = gwaResults.SNPResult(resultFile=srInput,
snpsds=snpsds,
phenotypeID=p_i)
# - filter results
result.negLogTransform()
res = SecondStageAnalysis.retrieveSecondRunSNPs(result, srTopQuantile,
srWindowSize)
snps = res['snps']
snpTypes = res['snpTypes']
chromosomes = res['chromosomes']
positions = res['positions']
marfs = res['marfs']
mafs = res['mafs']
#Calc kinship matrix:
print "Calculating kinship matrix."
sys.stdout.flush()
k_snps = kinshipSnpsDataset.getSnps()
#k_snps = _sampleSNPs_(k_snps,10000)
k = calcKinship(k_snps)
#print "K:",k
#Run Emma:
sys.stdout.write("Running Emma.\n")
sys.stdout.flush()
res = runEmma(phed, p_i, k, snps)
pvals = list(res["ps"])
genotype_var_perc = list(res["genotype_var_perc"])
#print zip(positions,pvals,genotype_var_perc)
sys.stdout.write("Writing results to file.\n")
sys.stdout.flush()
#Write results to file!
f = open(srOutput, "w")
f.write(
"Chromosome,position,p-value,marf,maf,genotype_var_perc,snpType,second_pos\n"
)
for (chr, pos, pval, marf, maf, gvp,
snpType) in zip(chromosomes, positions, pvals, marfs, mafs,
genotype_var_perc, snpTypes):
f.write(
str(chr) + "," + str(pos[0]) + "," + str(pval[0]) + "," +
str(marf) + "," + str(maf) + "," + str(gvp[0]) + "," +
str(snpType) + "," + str(pos[1]) + "\n")
f.close()
def _secondRun_(srOutput,srInput,srTopQuantile,srWindowSize,snpsds,phed,p_i,kinshipSnpsDataset):
"""
Sets up a second run, using and/or on SNPs.
"""
print "Preparing a second stage analyzis."
sys.stdout.flush()
#Collect top n% SNPs.
print "Loading first stage results."
sys.stdout.flush()
result = gwaResults.SNPResult(resultFile=srInput,snpsds=snpsds,phenotypeID=p_i)
# - filter results
result.negLogTransform()
res = SecondStageAnalysis.retrieveSecondRunSNPs(result,srTopQuantile,srWindowSize)
snps = res['snps']
snpTypes = res['snpTypes']
chromosomes = res['chromosomes']
positions = res['positions']
marfs = res['marfs']
mafs = res['mafs']
#Calc kinship matrix:
print "Calculating kinship matrix."
sys.stdout.flush()
k_snps = kinshipSnpsDataset.getSnps()
#k_snps = _sampleSNPs_(k_snps,10000)
k = calcKinship(k_snps)
#print "K:",k
#Run Emma:
sys.stdout.write( "Running Emma.\n")
sys.stdout.flush()
res = runEmma(phed,p_i,k,snps)
pvals = list(res["ps"])
genotype_var_perc = list(res["genotype_var_perc"])
#print zip(positions,pvals,genotype_var_perc)
sys.stdout.write("Writing results to file.\n")
sys.stdout.flush()
#Write results to file!
f = open(srOutput,"w")
f.write("Chromosome,position,p-value,marf,maf,genotype_var_perc,snpType,second_pos\n")
for (chr,pos,pval,marf,maf,gvp,snpType) in zip(chromosomes,positions,pvals,marfs,mafs,genotype_var_perc,snpTypes):
f.write(str(chr)+","+str(pos[0])+","+str(pval[0])+","+str(marf)+","+str(maf)+","+str(gvp[0])+","+str(snpType)+","+str(pos[1])+"\n")
f.close()
def _secondRun_(srOutput,srInput,srTopQuantile,srWindowSize,snpsds,phed,p_i,binary = False):
"""
Sets up a second run, using and/or on SNPs.
"""
print "Preparing a second stage analyzis."
sys.stdout.flush()
#Collect top n% SNPs.
print "Loading first stage results."
sys.stdout.flush()
result = gwaResults.SNPResult(resultFile=srInput,snpsds=snpsds,phenotypeID=p_i)
# - filter results
result.negLogTransform()
res = SecondStageAnalysis.retrieveSecondRunSNPs(result,srTopQuantile,srWindowSize)
snps = res['snps']
snpTypes = res['snpTypes']
chromosomes = res['chromosomes']
positions = res['positions']
marfs = res['marfs']
mafs = res['mafs']
if binary:
#Run Fisher's Exact Test:
sys.stdout.write( "Running Fisher's exact test (in R).\n")
sys.stdout.flush()
phenVals = phed.getPhenVals(p_i)
pvals = run_fet(snps,phenVals)
#print zip(positions,pvals,genotype_var_perc)
else:
#Run Emma:
sys.stdout.write( "Running Kruskal-Wallis (in R).\n")
sys.stdout.flush()
phenVals = phed.getPhenVals(p_i)
pvals = run_kw(snps,phenVals)
#print zip(positions,pvals,genotype_var_perc)
sys.stdout.write("Writing results to file.\n")
sys.stdout.flush()
#Write results to file!
f = open(srOutput,"w")
f.write("Chromosome,position,p-value,marf,maf,snpType,second_pos\n")
for (chr,pos,pval,marf,maf,snpType) in zip(chromosomes,positions,pvals,marfs,mafs,snpTypes):
f.write(str(chr)+","+str(pos[0])+","+str(pval)+","+str(marf)+","+str(maf)+","+str(snpType)+","+str(pos[1])+"\n")
f.close()
