| OMNI_individual {GBJ} | R Documentation |
omni_individual.R
Description
Computes the omnibus test statistic combining GBJ, GHC, minP, and SKAT. This version of the function assumes you have the individual factor data (i.e. genotypes) for each subject. If you only have summary statistics, use omni_ss(). You WILL NOT be able to use this function unless you have also loaded the SKAT package (install.packages("SKAT"); library(SKAT)).
Usage
OMNI_individual(null_model, factor_matrix, link_function, num_boots = 100)
Arguments
null_model |
An R regression model fitted using glm(). Do not use lm(), even for linear regression! |
factor_matrix |
An n*d matrix with each factor (i.e. each SNP) as one column. There should be no missing data. |
link_function |
Either "linear" or "logit" or "log". |
num_boots |
Number of bootstrap repetitions to find correlation matrix of set-based statistics. |
Value
A list with the elements:
OMNI |
The observed omnibus test statistic. |
OMNI_pvalue |
The p-value of the OMNI test |
err_code |
Sometimes if your p-value is very small (< 1*10^(-10)), R may run into numerical issues. This message will alert you if such a situation occurs. |
Examples
factor_matrix <- matrix(data=rbinom(n=1000, size=2, prob=0.3), ncol=5)
Y <- rnorm(n=200)
null_mod <- glm(Y ~ 1)
OMNI_individual(null_model=null_mod, factor_matrix=factor_matrix,
link_function='linear', num_boots=5)