RDA Shrunken Gene List Function

Description

A function that returns the shrunken gene (variable) names by RDA for a particular (alpha, delta) combination.

Usage

1
2
genelist.rda(x, y, alpha, delta, prior=table(y)/length(y),
             gnames=NULL, regularization="S")

Arguments

x

The training data set for which you want to obtain the shrunken gene list. It must be a numerical matrix. The columns are sample observations and the rows are variables.

y

The class labels for the columns of 'x'.

prior

A numerical vector that gives the prior proportion of each class. By default, it is set to be the sample frequencies unless users want to specify a different one.

alpha

A single regularization value for alpha. Users must supply this option.

delta

A threshold value for delta. Users must supply this option.

gnames

A character vector that specifies the names of the variables of the training data set 'x'. By default, it is set to be NULL and the function uses either the row names of 'x' (if it exists) or the row index 1:nrow(x). Users can provide their customized gene name list. However, the length of the name vector must be the same as the number of rows of 'x'.

regularization

The type of regularization. It is either 'S' or 'R'. The default value is 'S'.

Details

genelist.rda will return a vector of names for those shrunken genes by RDA for a particular (alpha, delta).

Value

A character vector of the names of the shrunken genes.

Author(s)

Yaqian Guo, Trevor Hastie and Robert Tibshirani

References

Guo, Y. et al. (2004) Regularized Discriminant Analysis and Its Application in Microarrays, Technical Report, Department of Statistics, Stanford University.

Examples

1
2
3
data(colon)
colon.x <- t(colon.x)
genenames <- genelist.rda(colon.x, colon.y, alpha=0.1, delta=0.3)