def loadData(bfile,
extractSim,
phenoFile,
missingPhenotype='-9',
loadSNPs=False,
standardize=True):
bed = Bed(bfile)
if (extractSim is not None):
f = open(extractSim)
csvReader = csv.reader(f)
extractSnpsSet = set([])
for l in csvReader:
extractSnpsSet.add(l[0])
f.close()
keepSnpsInds = [
i for i in xrange(bed.sid.shape[0]) if bed.sid[i] in extractSnpsSet
]
bed = bed[:, keepSnpsInds]
phe = None
if (phenoFile is not None):
bed, phe = loadPheno(bed, phenoFile, missingPhenotype)
if (loadSNPs):
bed = bed.read()
if (standardize): bed = bed.standardize()
return bed, phe
def load_data(snp_file, pheno_file, covar_file):
# Load SNP data
snp_reader = Bed(snp_file)
# Load phenotype
pheno = pysnptools.util.pheno.loadPhen(pheno_file)
# Load covariates
if covar_file is not None:
covar = pysnptools.util.pheno.loadPhen(covar_file)
snp_reader, pheno, covar = srutil.intersect_apply([snp_reader, pheno, covar])
covar = covar['vals']
else:
snp_reader, pheno = srutil.intersect_apply([snp_reader, pheno])
covar = None
snp_data = snp_reader.read().standardize()
Y = pheno['vals']
Y -= Y.mean(0)
Y /= Y.std(0)
X = 1./np.sqrt((snp_data.val**2).sum() / float(snp_data.iid_count)) * snp_data.val
K = np.dot(X, X.T) # TODO use symmetric dot to speed this up
assert np.all(pheno['iid'] == snp_data.iid), "the samples are not sorted"
return snp_data, pheno, covar, X, Y, K
def write_grm(K, out_file, num_snps=500000, K_index=None, bed_nold_file=None):
# fill GRM columns 0,1,3
n = K.shape[0]
grm = np.zeros((n * (n - 1) / 2 + n, 4))
tril_indices = np.tril_indices(n)
grm[:, 0] = tril_indices[0] + 1
grm[:, 1] = tril_indices[1] + 1
grm[:, 2] = num_snps
grm[:, 3] = K[tril_indices]
# fill #non-missing SNPs columns
if bed_nold_file is not None:
assert K_index is not None, 'K_index not provided'
from pysnptools.snpreader.bed import Bed
bed_nold = Bed(bed_nold_file, count_A1=True).read()
notNan = (~np.isnan(bed_nold.val)).astype(np.float)
notNan_K = notNan.dot(notNan.T)
id2ind = dict([])
for ind_i, ind in enumerate(bed_nold.iid[:, 1].astype(np.int)):
id2ind[ind] = ind_i
tril_indices_nold = [None, None]
tril_indices_nold[0] = [id2ind[K_index[ind]] for ind in tril_indices[0]]
tril_indices_nold[1] = [id2ind[K_index[ind]] for ind in tril_indices[1]]
grm[:, 2] = notNan_K[tril_indices_nold]
pd_grm = pd.DataFrame(grm, columns=['i', 'j', 'num_SNPs', 'K'])
pd_grm['i'] = pd_grm['i'].astype(np.int)
pd_grm['j'] = pd_grm['j'].astype(np.int)
pd_grm.to_csv(out_file, compression='gzip', header=False, index=False, sep='\t', float_format='%0.6e')
def loadData(bfile, extractSim, phenoFile, missingPhenotype='-9', loadSNPs=False, standardize=True): bed = Bed(bfile) if (extractSim is not None): f = open(extractSim) csvReader = csv.reader(f) extractSnpsSet = set([]) for l in csvReader: extractSnpsSet.add(l[0]) f.close() keepSnpsInds = [i for i in xrange(bed.sid.shape[0]) if bed.sid[i] in extractSnpsSet] bed = bed[:, keepSnpsInds] phe = None if (phenoFile is not None): bed, phe = loadPheno(bed, phenoFile, missingPhenotype) if (loadSNPs): bed = bed.read() if (standardize): bed = bed.standardize() return bed, phe
def _snps_fixup(snp_input, iid_if_none=None,count_A1=None):
if isinstance(snp_input, str):
return Bed(snp_input,count_A1=count_A1)
if isinstance(snp_input, dict):
return SnpData(iid=snp_input['iid'],sid=snp_input['header'],val=snp_input['vals'])
if snp_input is None:
assert iid_if_none is not None, "snp_input cannot be None here"
return SnpData(iid_if_none, sid=np.empty((0),dtype='str'), val=np.empty((len(iid_if_none),0)),pos=np.empty((0,3)),name="") #todo: make a static factory method on SnpData
return snp_input
def getChromosome(bfile, chrom):
bed = Bed(bfile)
indsToKeep = (bed.pos[:, 0] == chrom)
bed = bed[:, indsToKeep]
return bed.read().standardize()
def _fixupBed(bed):
if isinstance(bed, str):
return Bed(bed).read().standardize()
else:
return bed
def getChromosome(bfile, chrom): bed = Bed(bfile) indsToKeep = (bed.pos[:,0] == chrom) bed = bed[:, indsToKeep] return bed.read().standardize()
def getExcludedChromosome(bfile, chrom):
bed = Bed(bfile)
indsToKeep = bed.pos[:, 0] != chrom
bed = bed[:, indsToKeep]
return bed.read().standardize()
import pandas as pd
import numpy as np
import leap.leapUtils as leapUtils
import leap.leapMain as leapMain
from pysnptools.snpreader.bed import Bed
#Define analysis data
bfile = 'dataset1/dataset1'
phenoFile = bfile + '.phe'
chromosomes = range(1, 11)
prevalence = 0.001
#Find individuals to exclude to eliminate relatedness (kinship coeff > 0.05)
bed = Bed(bfile).read().standardize()
indsToKeep = leapUtils.findRelated(bed, cutoff=0.05)
#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)
def getExcludedChromosome(bfile, chrom):
bed = Bed(bfile, count_A1=True)
indsToKeep = (bed.pos[:, 0] != chrom)
bed = bed[:, indsToKeep]
return bed.read().standardize()
