Description Usage Arguments Details Value Note Author(s) Examples
Estimate and test the difference in SNP effect the trait between two strata.
1 |
beta |
a p by s matrix of estimated effects of each of p SNPs on the trait in each of 2 strata. |
var |
a p by 2 matrix of variances of the estimated effects of each of p SNPs on the trait in each of 2 strata. |
cov |
a p by 1 matrix of a vector of length p of covairances between the estimated effects of each of p SNPs on the trait in each of 2 strata. |
For each SNP, estimates the difference in SNP effect between strata, and tests the null hypothesis that this effect is zero.
A matrix with column corresponding to beta = the estimated effect difference (equivalently, the interaction effect), var = the variance of the estimated beta, test.stat = the test.statistics and pval = the p-value of the test.
This function is used by stratLMMTest when two strata are analyzed.
Tamar Sofer
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 | ### create a small data set with the right formating:
require(mvtnorm)
n.strat <- 5
n.est <- 5
## create a positive definite matrix to be a covariance matrix
sigma <- matrix(rnorm(n.strat^2), n.strat, n.strat)
sigma <- t(sigma)
effects <- rmvnorm(n.est, c(2,2,3,3,0), sigma)
colnames(effects) <- paste0("Beta.s", 1:n.strat)
vars <- matrix(rep(diag(sigma), n.est), nrow = n.est, ncol = n.strat, byrow = T)
colnames(vars) <- paste0("var.s", 1:n.strat)
covs <- matrix(rep(sigma[upper.tri(sigma)], n.est), nrow = n.est, byrow = T)
colnames(covs) <- c(paste0("cov.s1:", paste0("s", 2:n.strat)), paste0("cov.s2:", paste0("s", 3:n.strat)), paste0("cov.s3:", paste0("s", 4:n.strat)) , "cov.s4:s5")
## this will give an error:
# test2levelInteraction(effects, vars, covs)
### error! there are more than 2 strata
test2levelInteraction(effects[,1:2], vars[,1:2], covs[,1])
test2levelInteraction(effects[,4:5], vars[,4:5], covs[,grep("cov.s4:s5", colnames(covs))])
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.