Skip to content
Diptavo Dutta edited this page Dec 8, 2017 · 3 revisions

Multi-SKAT R-package

This package produces kernel Regression based rare-variant association tests for Multiple phenotypes. The functions aggregate variant-phenotype score statistic in a particular region/gene and computes corresponding p-values efficiently.

Overview

In this vignette we display an elementary workflow to obtain the association test results corresponding to different Multi-SKAT tests (omnibus and with pre-specified kernel)

Unrelated individuals

Multi-SKAT with pre-specified kernels

An example dataset MultiSKAT.example.data has a genotype matrix Genotypes of 5000 individuals and 56 SNPs, a phenotype matrix Phenotypes of 5 continuous phenotypes on those individuals, and a covariates vector Cov denoting intercept.

The first step is to create a null model using MultiSKAT_NULL function (for unrelated individuals) with the phenotype matrix and covariate matrix. Subsequently, MultiSKAT function is used with appropriate kernel to obtain the association p-value.

Clone this wiki locally