empirical.controls {sva} | R Documentation |
This function uses the iteratively reweighted surrogate variable analysis approach to estimate the probability that each gene is an empirical control.
empirical.controls(dat, mod, mod0 = NULL, n.sv, B = 5, type = c("norm", "counts"))
dat |
The transformed data matrix with the variables in rows and samples in columns |
mod |
The model matrix being used to fit the data |
mod0 |
The null model being compared when fitting the data |
n.sv |
The number of surogate variables to estimate |
B |
The number of iterations of the irwsva algorithm to perform |
type |
If type is norm then standard irwsva is applied, if type is counts, then the moderated log transform is applied first |
pcontrol A vector of probabilites that each gene is a control.
library(bladderbatch) data(bladderdata) dat <- bladderEset[1:5000,] pheno = pData(dat) edata = exprs(dat) mod = model.matrix(~as.factor(cancer), data=pheno) n.sv = num.sv(edata,mod,method="leek") pcontrol <- empirical.controls(edata,mod,mod0=NULL,n.sv=n.sv,type="norm")