aGE {aGE}R Documentation

aGE interaction test

Description

aGE interaction test

Usage

aGE(Y, G, cov = NULL, model = c("gaussian", "binomial"),
  pow = c(1:6), n.perm = 1000, method = "Simulation", nonparaE = F,
  DF = 10, stepwise = T)

Arguments

Y

a numeric vector of phenotype values

G

a matrix for all RVs in the test gene or genomic region. The order of rows must match the order of Y. Missing is imputed as 0.

cov

a matrix with first column as the environmental variable to be tested. The order of rows must match the order of Y.

model

"binomial" for binary traits or "gaussian" for quantitative traits.

pow

Gamma set used to build a family of tests, default=c(1:6) for rare variants

n.perm

number of simulation to calculate the p-values, default=1000. Can increase to higher value depending on the signficiance level.

method

only have one option: "Simulation", also called Monte Carlo Method.

nonparaE

"T": use cubic splines for the environmental variable to fit the model; "F": use a linear function of the environmental variable to fit the model

DF

degree of freedom to use in the cubic splines, default=10. This option only works when nonparaE is set to "T"

stepwise

an option to speed up the simulation procedure for large n.perm number in real-data application. Up to $n.perm=10^8$

Value

p-values

Examples

{
 set.seed(12345)
 phenotype <- c(rep(1,50),rep(0,50))
 genotype <- data.frame(g1=sample(c(rep(1,10),rep(0,90))),g2=sample(c(rep(1,5), rep(0,95))))
 covariates <- data.frame(Envir=rnorm(100), Age=rnorm(100,60,5))
 exD <- list(Y=phenotype, G=genotype, X=covariates)
 aGE(Y=exD$Y, G=exD$G, cov=exD$X, model='binomial', nonparaE=FALSE, stepwise=FALSE)
 }

[Package aGE version 0.0.9 Index]