def work_sequence(self):
# is it OK to do the intersect and the linear regression 23 extra times?
# clear
G, y, snp_name, _ = load_intersect(self.snp_reader, self.pheno_fn)
# compute linear regression
_, p_values_lin = f_regression(G, y, center=True)
# set up empty return structures
#self.rs = snp_name
#self.p_values = -np.ones(len(snp_name))
# get chr names/id
chr_ids = self.snp_reader.pos[:,0]
#self.pos = self.snp_reader.pos
#loco = [[range(0,5000), range(5000,10000)]]
loco = LeaveOneChromosomeOut(chr_ids, indices=True)
if len(loco) is not self.chrom_count : raise Exception("The snp reader has {0} chromosome, not {1} as specified".format(len(loco),self.chrom_count))
for i, (train_snp_idx, test_snp_idx) in enumerate(loco):
if i == 0:
result = {"p_values":-np.ones(len(snp_name)),
"p_values_lin": p_values_lin,
"rs":snp_name,
"pos":self.snp_reader.pos}
else:
result = None
yield lambda i=i, train_snp_idx=train_snp_idx,test_snp_idx=test_snp_idx,result=result,G=G,y=y: self.dowork(i,train_snp_idx,test_snp_idx,result,G,y) # the 'i=i',etc is need to get around a strangeness in Python
def test_results_identical_with_fastlmmc(self):
"""
make sure gwas yields same results as fastlmmC
"""
currentFolder = os.path.dirname(os.path.realpath(__file__))
#prefix = r"C:\Users\chwidmer\Documents\Projects\sandbox\data\test"
#bed_fn = prefix + "/jax_gt.up.filt.M"
#dat_fn = prefix + "/jax_M_expression.1-18.dat"
#pheno_fn = prefix + "/jax_M_expression.19.phe.txt"
bed_fn = os.path.join(currentFolder, "../../feature_selection/examples/toydata")
pheno_fn = os.path.join(currentFolder, "../../feature_selection/examples/toydata.phe")
#prefix = "../../../tests\datasets\mouse"
#bed_fn = os.path.join(prefix, "alldata")
#pheno_fn = os.path.join(prefix, "pheno.txt")
snp_reader = Bed(bed_fn)
G, y, _, _ = load_intersect(snp_reader, pheno_fn)
snp_pos = snp_reader.rs
idx_sim = range(0, 5000)
idx_test = range(5000, 10000)
snp_pos_sim = snp_pos[idx_sim]
snp_pos_test = snp_pos[idx_test]
G_chr1, G_chr2 = G[:,idx_sim], G[:,idx_test]
delta = 1.0
###################################
# REML IN lmm.py is BROKEN!!
# we compare REML=False in lmm.py to fastlmmc
REML = False
gwas_c_reml = GwasTest(bed_fn, pheno_fn, snp_pos_sim, snp_pos_test, delta, REML=REML)
gwas_c_reml.run_gwas()
gwas = GwasPrototype(G_chr1, G_chr2, y, delta, REML=False)
gwas.run_gwas()
# check p-values in log-space!
np.testing.assert_array_almost_equal(np.log(gwas.p_values), np.log(gwas_c_reml.p_values), decimal=3)
if False:
import pylab
pylab.plot(np.log(gwas_c_reml.p_values), np.log(gwas_f.p_values_F), "x")
pylab.plot(range(-66,0,1), range(-66,0,1))
pylab.show()
# we compare lmm_cov.py to fastlmmc with REML=False
gwas_c = GwasTest(bed_fn, pheno_fn, snp_pos_sim, snp_pos_test, delta, REML=True)
gwas_c.run_gwas()
gwas_f = FastGwas(G_chr1, G_chr2, y, delta, findh2=False)
gwas_f.run_gwas()
np.testing.assert_array_almost_equal(np.log(gwas_c.p_values), np.log(gwas_f.p_values_F), decimal=2)
# additional testing code for the new wrapper functions
# Fix delta
from pysnptools.snpreader import Bed as BedSnpReader
from fastlmm.association.single_snp import single_snp
snpreader = BedSnpReader(bed_fn,count_A1=False)
frame = single_snp(test_snps=snpreader[:,idx_test], pheno=pheno_fn, G0=snpreader[:,idx_sim],h2=1.0/(delta+1.0),leave_out_one_chrom=False,count_A1=False)
sid_list,pvalue_list = frame['SNP'].values,frame['PValue'].values
np.testing.assert_allclose(gwas_f.sorted_p_values_F, pvalue_list, rtol=1e-10)
p_vals_by_genomic_pos = frame.sort_values(["Chr", "ChrPos"])["PValue"].tolist()
np.testing.assert_allclose(gwas_c_reml.p_values, p_vals_by_genomic_pos, rtol=.1)
np.testing.assert_allclose(gwas_c_reml.p_values, gwas_f.p_values_F, rtol=.1)
np.testing.assert_allclose(gwas_f.sorted_p_values_F, gwas_c_reml.sorted_p_values, rtol=.1)
# Search over delta
gwas_c_reml_search = GwasTest(bed_fn, pheno_fn, snp_pos_sim, snp_pos_test, delta=None, REML=True)
gwas_c_reml_search.run_gwas()
frame_search = single_snp(test_snps=snpreader[:,idx_test], pheno=pheno_fn, G0=snpreader[:,idx_sim],h2=None,leave_out_one_chrom=False,count_A1=False)
_,pvalue_list_search = frame_search['SNP'].values,frame_search['PValue'].values
p_vals_by_genomic_pos = frame_search.sort_values(["Chr", "ChrPos"])["PValue"].tolist()
np.testing.assert_allclose(gwas_c_reml_search.p_values, p_vals_by_genomic_pos, rtol=.001)
np.testing.assert_allclose(gwas_c_reml_search.sorted_p_values, pvalue_list_search, rtol=.001)
def test_results_identical_with_fastlmmc(self):
"""
make sure gwas yields same results as fastlmmC
"""
currentFolder = os.path.dirname(os.path.realpath(__file__))
#prefix = r"C:\Users\chwidmer\Documents\Projects\sandbox\data\test"
#bed_fn = prefix + "/jax_gt.up.filt.M"
#dat_fn = prefix + "/jax_M_expression.1-18.dat"
#pheno_fn = prefix + "/jax_M_expression.19.phe.txt"
bed_fn = os.path.join(currentFolder, "../../feature_selection/examples/toydata")
pheno_fn = os.path.join(currentFolder, "../../feature_selection/examples/toydata.phe")
#prefix = "../../../tests\datasets\mouse"
#bed_fn = os.path.join(prefix, "alldata")
#pheno_fn = os.path.join(prefix, "pheno.txt")
snp_reader = Bed(bed_fn)
G, y, _, _ = load_intersect(snp_reader, pheno_fn)
snp_pos = snp_reader.rs
idx_sim = range(0, 5000)
idx_test = range(5000, 10000)
snp_pos_sim = snp_pos[idx_sim]
snp_pos_test = snp_pos[idx_test]
G_chr1, G_chr2 = G[:,idx_sim], G[:,idx_test]
delta = 1.0
###################################
# REML IN lmm.py is BROKEN!!
# we compare REML=False in lmm.py to fastlmmc
REML = False
gwas_c_reml = GwasTest(bed_fn, pheno_fn, snp_pos_sim, snp_pos_test, delta, REML=REML)
gwas_c_reml.run_gwas()
gwas = GwasPrototype(G_chr1, G_chr2, y, delta, REML=False)
gwas.run_gwas()
# check p-values in log-space!
np.testing.assert_array_almost_equal(np.log(gwas.p_values), np.log(gwas_c_reml.p_values), decimal=3)
if False:
import pylab
pylab.plot(np.log(gwas_c_reml.p_values), np.log(gwas_f.p_values_F), "x")
pylab.plot(range(-66,0,1), range(-66,0,1))
pylab.show()
# we compare lmm_cov.py to fastlmmc with REML=False
gwas_c = GwasTest(bed_fn, pheno_fn, snp_pos_sim, snp_pos_test, delta, REML=True)
gwas_c.run_gwas()
gwas_f = FastGwas(G_chr1, G_chr2, y, delta, findh2=False)
gwas_f.run_gwas()
np.testing.assert_array_almost_equal(np.log(gwas_c.p_values), np.log(gwas_f.p_values_F), decimal=2)
# additional testing code for the new wrapper functions
# Fix delta
from pysnptools.snpreader import Bed as BedSnpReader
from fastlmm.association.single_snp import single_snp
snpreader = BedSnpReader(bed_fn,count_A1=False)
frame = single_snp(test_snps=snpreader[:,idx_test], pheno=pheno_fn, G0=snpreader[:,idx_sim],h2=1.0/(delta+1.0),leave_out_one_chrom=False,count_A1=False)
sid_list,pvalue_list = frame['SNP'].values,frame['PValue'].values
np.testing.assert_allclose(gwas_f.sorted_p_values_F, pvalue_list, rtol=1e-10)
p_vals_by_genomic_pos = frame.sort_values(["Chr", "ChrPos"])["PValue"].tolist()
np.testing.assert_allclose(gwas_c_reml.p_values, p_vals_by_genomic_pos, rtol=.1)
np.testing.assert_allclose(gwas_c_reml.p_values, gwas_f.p_values_F, rtol=.1)
np.testing.assert_allclose(gwas_f.sorted_p_values_F, gwas_c_reml.sorted_p_values, rtol=.1)
# Search over delta
gwas_c_reml_search = GwasTest(bed_fn, pheno_fn, snp_pos_sim, snp_pos_test, delta=None, REML=True)
gwas_c_reml_search.run_gwas()
frame_search = single_snp(test_snps=snpreader[:,idx_test], pheno=pheno_fn, G0=snpreader[:,idx_sim],h2=None,leave_out_one_chrom=False,count_A1=False)
_,pvalue_list_search = frame_search['SNP'].values,frame_search['PValue'].values
p_vals_by_genomic_pos = frame_search.sort_values(["Chr", "ChrPos"])["PValue"].tolist()
np.testing.assert_allclose(gwas_c_reml_search.p_values, p_vals_by_genomic_pos, rtol=.001)
np.testing.assert_allclose(gwas_c_reml_search.sorted_p_values, pvalue_list_search, rtol=.001)