def setUpClass(self):
from fastlmm.util.util import create_directory_if_necessary
create_directory_if_necessary(self.tempout_dir, isfile=False)
self.pythonpath = os.path.abspath(os.path.join(os.path.dirname(os.path.realpath(__file__))))
self.bedbase = os.path.join(self.pythonpath, 'dataset1/dataset1')
self.phen_fn = os.path.join(self.pythonpath, 'dataset1/dataset1.phe')
#Create eigendecompositions
logging.info("Creating eigendecomposition files")
for i in xrange(1,11):
output_file = os.path.abspath(os.path.join(self.tempout_dir, 'dataset1_nochr{}.npz'.format(i)))
extractSim = 'dataset1/extracts/nochr{0}_extract.txt'.format(i)
bed, _ = leapUtils.loadData(self.bedbase, extractSim, self.phen_fn, loadSNPs=True)
leapMain.eigenDecompose(bed, output_file)
#Iterate over each chromosome
frame_list = []
for chrom in chromosomes:
print()
print('Analyzing chromosome', chrom, '...')
#Create a bed object excluding SNPs from the current chromosome
bedExclude = leapUtils.getExcludedChromosome(bfile, chrom)
#Create a bed object including only SNPs from the current chromosome
bedTest = leapUtils.getChromosome(bfile, chrom)
#Compute eigendecomposition for the data
eigenFile = 'temp_eigen.npz'
eigen = leapMain.eigenDecompose(bedExclude, outFile=eigenFile)
#compute heritability explained by this data
h2 = leapMain.calcH2(phenoFile,
prevalence,
eigen,
keepArr=indsToKeep,
h2coeff=1.0)
#Compute liabilities explained by this data
liabs = leapMain.probit(bedExclude,
phenoFile,
h2,
prevalence,
eigen,
keepArr=indsToKeep)
#Iterate over each chromosome
frame_list = []
for chrom in chromosomes:
print
print 'Analyzing chromosome', chrom, '...'
#Create a bed object excluding SNPs from the current chromosome
bedExclude = leapUtils.getExcludedChromosome(bfile, chrom)
#Create a bed object including only SNPs from the current chromosome
bedTest = leapUtils.getChromosome(bfile, chrom)
#Compute eigendecomposition for the data
eigenFile = 'temp_eigen.npz'
eigen = leapMain.eigenDecompose(bedExclude, outFile=eigenFile)
#compute heritability explained by this data
h2 = leapMain.calcH2(phenoFile, prevalence, eigen, keepArr=indsToKeep)
#Compute liabilities explained by this data
liabs = leapMain.probit(bedExclude, phenoFile, h2, prevalence, eigen, keepArr=indsToKeep)
#perform GWAS, using the liabilities as the observed phenotypes
results_df = leapMain.leapGwas(bedExclude, bedTest, liabs, h2)
frame_list.append(results_df)
#Join together the results of all chromosomes, and print the top ranking SNPs
frame = pd.concat(frame_list)
frame.sort("PValue", inplace=True)
frame.index = np.arange(len(frame))
