Meta_Null_Model: Get parameters and residuals from H0

In MetaSKAT: Meta Analysis for SNP-Set (Sequence) Kernel Association Test

Get parameters and residuals from H0

Description

Compute model parameters and residuals under the null model (H0) of no associations. It can be used only when individual level data are available.

Usage

Meta_Null_Model(y.list, x.list, n.cohort, out_type="C",  n.Resampling=0)
Meta_Null_Model_EmmaX(y.list, x.list, n.cohort, K.list=NULL, Kin.File.list=NULL)

 

Arguments

y.list	a list object for phenotypes. Each element should be a vector of phenotypes. If you have 3 cohorts, it should have 3 elements.
x.list	a list object for covariates. Each element should be a vector or a matrix of covariates. If there are 3 cohorts, it should have 3 elements. If there are no covariates to adjust for, the element should be "intercept". See the examples.
n.cohort	a numeric value of the number of cohort.
out_type	an indicator for the outcome type. "C" for continuous outcomes and "D" for dichotomous outcomes.
n.Resampling	internal use only.
K.list	a list object of the kinship matrices. If K.list=NULL, the function reads files in Kin.File.list to obtain kinship matrices.
Kin.File.list	a list object of emmax-kin output file names.

Value

It returns an object that has model parameters and residuals. The returned object will be used to run MetaSKAT_wZ.

Author(s)

Seunggeun Lee

Examples

data(Example)
attach(Example)

#############################################################
#	Compute a p-value of the first gene

obj<-Meta_Null_Model(y.list, x.list, n.cohort=3, out_type="D")
MetaSKAT_wZ(Z.list[[1]], obj)$p.value

#############################################################
#	If you want to use the intercept-only model for the 2nd cohort

x.list[[2]]<-"intercept"
obj<-Meta_Null_Model(y.list, x.list, n.cohort=3, out_type="D")
MetaSKAT_wZ(Z.list[[1]], obj)$p.value

MetaSKAT documentation built on Oct. 8, 2024, 5:07 p.m.