KS.prelim {KnockoffScreen}

Preliminary data management for KnockoffScreen

Description

This function does the preliminary data management and fit the model under null hypothesis. The output will be passed to the other functions.

Usage

KS.prelim(Y, X=NULL, id=NULL, out_type="C")

Arguments

Y

The outcome variable, an n*1 matrix where n is the total number of observations

X

An n*d covariates matrix where d is the total number of covariates.

id

The subject id. This is used to match phenotype with genotype. The default is NULL, where the matched phenotype and genotype matrices are assumed.

out_type

Type of outcome variable. Can be either "C" for continuous or "D" for dichotomous. The default is "C".

Value

It returns a list that will be passed to function KS.chr().

Examples

library(KnockoffScreen)

# load example vcf file from package "seqminer"
vcf.filename = system.file("vcf/1000g.phase1.20110521.CFH.var.anno.vcf.gz", package = "seqminer")

## this is how the actual genotype matrix from package "seqminer" looks like
example.G <- t(readVCFToMatrixByRange(vcf.filename, "1:196621007-196716634",annoType='')[[1]])

# simulated outcomes, covariates and inidividual id.
Y<-as.matrix(rnorm(nrow(example.G),0,1))
X<-as.matrix(rnorm(nrow(example.G),0,1))
id<-rownames(example.G)

# fit null model
result.prelim<-KS.prelim(Y,X=X,id=id,out_type="C")