num.sv: A function for calculating the number of surrogate variables to estimate in a model
Description
This function estimates the number of surrogate variables that should be included
in a differential expression model. The default approach is based on a permutation
procedure originally prooposed by Buja and Eyuboglu 1992. The function also provides
an interface to the asymptotic approach proposed by Leek 2011 Biometrics.
Usage
num.sv(dat, mod, method = c("be", "leek"), vfilter = NULL, B = 20, seed = NULL)
Arguments
dat
The transformed data matrix with the variables in rows and samples in columns
mod
The model matrix being used to fit the data
method
One of "be" or "leek" as described in the details section
vfilter
You may choose to filter to the vfilter most variable rows before performing the analysis
B
The number of permutaitons to use if method = "be"
seed
Set a seed when using the permutation approach
Value
n.sv The number of surrogate variables to use in the sva software