famSKAT_RC: famSKAT_RC

Description Usage Arguments Details Value Examples

View source: R/famSKAT_RC.R

Description

FamSKAT-RC is a family-based association kernel test for both rare and common variants.

Usage

1
2
famSKAT_RC (PHENO, genotypes, id, fullkins, covariates=NULL, sqrtweights_c,
            sqrtweights_r, binomialimpute=FALSE, acc=NULL, maf, phi)

Arguments

PHENO

The vector of the phenotype values. The missing values must be represented by NA.

genotypes

The genotype matrix. Its size should be N * P, where N is the number of individuals and P is the number of SNPs.

id

The vector of individual IDs to be included in the test. These IDs are present in the kinship matrix.

fullkins

The full kinship matrix that contains all individuals.

covariates

A matrix of possible covariates.

sqrtweights_c

The weight function to be assigned for common variants. An example is: function(maf) ifelse(maf>0, dbeta(maf,0.5,0.5), 0).

sqrtweights_r

The weight function to be assigned for rare variants. An example is: function(maf) ifelse(maf>0, dbeta(maf,1,25), 0).

binomialimpute

If TRUE, then impute missing genotypes using a binomial distribution (e.g. rbinom(10, 2, MAF), if 10 genotypes are missing).

acc

The accuracy used in the Davies approximation. For example, acc= 1e-06.

maf

A MAF threshold used to define rare and common variants (e.g. maf=0.01).

phi

This parameter indicates the contribution portion of rare variants. For example, a value of phi=0.5 means that the contribution of rare and common variants to the test is equaland a value of phi=1 means that only rare variants contribute to the test. You can vary the phi values as you wish and you can also provide a grid of different values (e.g. phi = c(0,0.2,0.5,0.9)). In this case, four p-values will be obtained, one for each value of phi.

Details

FamSKAT-RC is a family-based association kernel test for both rare and common variants. This test is general and several special cases are known as other methods: famSKAT, which only focuses on rare variants in family-based data, SKAT, which focuses on rare variants in population-based data (unrelated individuals), and SKAT-RC, which focuses on both rare and common variants in population-based data. When one applies famSKAT-RC and sets the value of phi to 1, famSKAT-RC becomes famSKAT. When one applies famSKAT-RC and set the value of phi to 1 and the kinship matrix to the identity matrix, famSKAT-RC becomes SKAT. When one applies famSKAT-RC and set the kinship matrix (fullkins) to the identity matrix (and phi is not equal to 1), famSKAT-RC becomes SKAT-RC.

Value

The test statistic p-value.

temp_pvalues

Some Description

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
library(kinship2)
sample.ped.geno <- process_data()
KIN = kinship(sample.ped.geno$IID, sample.ped.geno$FA, sample.ped.geno$MO)
IID = sample.ped.geno$IID
wuweights_r <- function(maf) ifelse(maf>0, dbeta(maf,1,25), 0)
wuweights_c <- function(maf) ifelse(maf>0, dbeta(maf,0.5,0.5), 0)
P_VALUES <- famSKAT_RC(PHENO=sample.ped.geno[,"Phenotype"],genotypes=as.matrix(
            sample.ped.geno[,7:ncol(sample.ped.geno)]), binomialimpute=TRUE,
            id=IID,fullkins=KIN,maf=0.05, sqrtweights_c=wuweights_c,
            sqrtweights_r=wuweights_r, phi = c(0,0.2,0.5,0.9))
print(P_VALUES)

famSKATRC documentation built on Nov. 17, 2017, 4:42 a.m.