SKAT: implements the sequence kernel association test for GWAS data
In kernelPSI: Post-Selection Inference for Nonlinear Variable Selection

Description Usage Arguments Details Value References Examples

The SKAT test is a quadratic test of association between a phenotype of interest and a genomic region. One of the main benefits of the SKAT test is the incorporation of nonlinear effects through the use of a kernel similarity matrix in the quadratic form. For instance, the identical-by-state (IBS) kernel which computes the number of identical alleles between two samples can be used.

1	SKAT(Y, K, sigma = 1)

`Y`	response vector
`K`	list of kernel similarity matrices. The sum kernel is used in the quadratic form.
`sigma`	standard deviation of the response Y

The null hypothesis in the SKAT test is the absence of effects of the SNPs within the region of interest and the outcome. Under the null, the distribution of the test statistic is a weighted sum of chi-square distributions whose quantiles are computed using the davies formula.

p-value of the SKAT test

Wu, M. C., Lee, S., Cai, T., Li, Y., Boehnke, M., & Lin, X. (2011). Rare-variant association testing for sequencing data with the sequence kernel association test. American Journal of Human Genetics, 89(1), 82–93.

n <- 30
p <- 20
K <- replicate(5, matrix(rnorm(n*p), nrow = n, ncol = p), simplify = FALSE)
K <-  sapply(K, function(X) return(X %*% t(X) / dim(X)[2]), simplify = FALSE)
Y <- rnorm(n)
SKAT(Y, K)