def update_associations(self, startTraitIdx=0, nTraits=np.inf):
"""
updating association scan
"""
self.assoc_updates = assoc_results.AssocResultsList()
focal_gene_prev = None
for focal_gene, snp_anchor, orth_gene in self.vstructures.iterator():
if focal_gene<startTraitIdx or startTraitIdx+nTraits<=focal_gene:
continue
if (focal_gene_prev is None) or (focal_gene!=focal_gene_prev):
print ".... Updating associations for gene %d"%(focal_gene)
focal_gene_prev = focal_gene
y_focal = self.phenoreader.getRows(focal_gene).T
y_orth = self.phenoreader.getRows(orth_gene).T
if y_focal.ndim==1: y_focal = y_focal[:,np.newaxis]
if y_orth.ndim==1: y_orth = y_orth[:,np.newaxis]
startSnpIdx = np.min(snp_anchor)
nSnps = np.max(snp_anchor) - startSnpIdx + 1
G_anchor = self.genoreader.loadSnpBlock(startSnpIdx, nSnps).T
G_anchor = G_anchor[:,snp_anchor-startSnpIdx]
pv, beta, var_snps, var_covs, var_genes = qtl_lr.test_lmm_lr_speed(G_anchor,y_focal, Z=y_orth,Kbg=self.K,Covs=self.Covs, S=self.S, U=self.U)
self.assoc_updates.add(pv,beta, var_snps, var_covs, var_genes, snp_anchor, focal_gene)
def update_associations(self, startTraitIdx=0, nTraits=np.inf):
"""
updating association scan
"""
self.assoc_updates = assoc_results.AssocResultsList()
focal_gene_prev = None
for focal_gene, snp_anchor, orth_gene in self.vstructures.iterator():
if focal_gene < startTraitIdx or startTraitIdx + nTraits <= focal_gene:
continue
if (focal_gene_prev is None) or (focal_gene != focal_gene_prev):
print ".... Updating associations for gene %d" % (focal_gene)
focal_gene_prev = focal_gene
y_focal = self.phenoreader.getRows(focal_gene).T
y_orth = self.phenoreader.getRows(orth_gene).T
if y_focal.ndim == 1:
y_focal = y_focal[:, np.newaxis]
if y_orth.ndim == 1:
y_orth = y_orth[:, np.newaxis]
startSnpIdx = np.min(snp_anchor)
nSnps = np.max(snp_anchor) - startSnpIdx + 1
G_anchor = self.genoreader.loadSnpBlock(startSnpIdx, nSnps).T
G_anchor = G_anchor[:, snp_anchor - startSnpIdx]
pv, beta, var_snps, var_covs, var_genes = qtl_lr.test_lmm_lr_speed(
G_anchor, y_focal, Z=y_orth, Kbg=self.K, Covs=self.Covs, S=self.S, U=self.U
)
self.assoc_updates.add(pv, beta, var_snps, var_covs, var_genes, snp_anchor, focal_gene)
def block_associations(self, startTraitIdx=0, nTraits=np.inf):
"""
blocking association scan by conditioning on the anchor gene
"""
self.assoc_updates = assoc_results.AssocResultsList()
focal_gene_prev = None
for focal_gene, snp_anchor, orth_gene, anchor_gene in self.vstructures.iterator(full=True):
if focal_gene < startTraitIdx or startTraitIdx + nTraits <= focal_gene:
continue
if (focal_gene_prev is None) or (focal_gene != focal_gene_prev):
print ".... Blocking associations for gene %d" % (focal_gene)
focal_gene_prev = focal_gene
y_focal = self.phenoreader.getRows(focal_gene).T
y_orth = self.phenoreader.getRows(orth_gene).T
if y_focal.ndim == 1:
y_focal = y_focal[:, np.newaxis]
if y_orth.ndim == 1:
y_orth = y_orth[:, np.newaxis]
startSnpIdx = np.min(snp_anchor)
nSnps = np.max(snp_anchor) - startSnpIdx + 1
G_anchor = self.genoreader.loadSnpBlock(startSnpIdx, nSnps).T
G_anchor = G_anchor[:, snp_anchor - startSnpIdx]
for _anchor_gene in np.unique(anchor_gene):
Covs = self.phenoreader.getRows([_anchor_gene]).T
if self.Covs is not None:
Covs = np.hstack((Covs, self.Covs))
pv, beta, var_snps, var_covs, var_genes = qtl_lr.test_lmm_lr_speed(
G_anchor[:, anchor_gene == _anchor_gene],
y_focal,
Z=y_orth,
Kbg=self.K,
Covs=Covs,
S=self.S,
U=self.U,
)
# pv, beta = qtl_lr.test_lmm_lr(G_anchor,y_focal, Z=y_orth,Kbg=self.K,Covs=self.Covs)
self.assoc_updates.add(
pv, beta, var_snps, var_covs, var_genes, snp_anchor[anchor_gene == _anchor_gene], focal_gene
)
def block_associations(self, startTraitIdx=0, nTraits=np.inf):
"""
blocking association scan by conditioning on the anchor gene
"""
self.assoc_updates = assoc_results.AssocResultsList()
focal_gene_prev = None
for focal_gene, snp_anchor, orth_gene,anchor_gene in self.vstructures.iterator(full=True):
if focal_gene<startTraitIdx or startTraitIdx+nTraits<=focal_gene:
continue
if (focal_gene_prev is None) or (focal_gene!=focal_gene_prev):
print ".... Blocking associations for gene %d"%(focal_gene)
focal_gene_prev = focal_gene
y_focal = self.phenoreader.getRows(focal_gene).T
y_orth = self.phenoreader.getRows(orth_gene).T
if y_focal.ndim==1: y_focal = y_focal[:,np.newaxis]
if y_orth.ndim==1: y_orth = y_orth[:,np.newaxis]
startSnpIdx = np.min(snp_anchor)
nSnps = np.max(snp_anchor) - startSnpIdx + 1
G_anchor = self.genoreader.loadSnpBlock(startSnpIdx, nSnps).T
G_anchor = G_anchor[:,snp_anchor-startSnpIdx]
for _anchor_gene in np.unique(anchor_gene):
Covs = self.phenoreader.getRows([_anchor_gene]).T
if self.Covs is not None:
Covs = np.hstack((Covs, self.Covs))
pv, beta, var_snps, var_covs, var_genes = qtl_lr.test_lmm_lr_speed(G_anchor[:,anchor_gene==_anchor_gene],y_focal, Z=y_orth,Kbg=self.K,Covs=Covs, S=self.S, U=self.U)
#pv, beta = qtl_lr.test_lmm_lr(G_anchor,y_focal, Z=y_orth,Kbg=self.K,Covs=self.Covs)
self.assoc_updates.add(pv,beta, var_snps, var_covs, var_genes, snp_anchor[anchor_gene==_anchor_gene], focal_gene)
