make.ref.model.matrices: Make Model Matrices for Random Effects Model

In Sage-Bionetworks/snm: Supervised Normalization of Microarrays

Description

Makes model matrices for random effects models

Usage

make.ref.model.matrices(snm.obj, exp)

Arguments

snm.obj	An object of class snm
exp	New environment to hold the lmer formula

Author(s)

Brig Mecham <[e-mail]> and John D. Storey <[e-mail]>

Examples

##---- Should be DIRECTLY executable !! ----
##-- ==>  Define data, use random,
##--	or do  help(data=index)  for the standard data sets.

## The function is currently defined as
function (snm.obj, exp) 
{
    F.mats <- list()
    spline.re.model <- ""
    for (i in 1:length(snm.obj$int.var)) {
        ide <- names(snm.obj$int.var)[i]
        spline.re.model <- paste(spline.re.model, "+ ( Bt", 1, 
            "-1 | ", ide, ")", sep = "")
        for (d in 2:snm.obj$spline.dim) {
            spline.re.model <- paste(spline.re.model, "+ ( Bt", 
                d, "-1 | ", ide, ")", sep = "")
        }
        F.mats[[i]] <- model.matrix(~-1 + snm.obj$int.var[, 
            i])
    }
    if (dim(snm.obj$adj.var)[2] > 0) {
        ref.model <- paste( "probes+", paste("probes:", colnames(snm.obj$adj.var)[-1],
            collapse = "+"),sep="+" )
    }
    else {
        ref.model <- "probes + "
    }
    ref.model <- as.formula(paste("y ~ -1 +", ref.model, spline.re.model), 
        env = exp)
    names(F.mats) <- names(snm.obj$int.var)
    list(ZF = ref.model, F.mats = F.mats)
  }

Sage-Bionetworks/snm documentation built on May 9, 2019, 12:14 p.m.