supclust (version 1.1-1)

sign.flip: Sign-flipping of Predictor Variables to Obtain Equal Polarity

Description

Computes the score for each predictor variable (gene) in the x-Matrix, and multiplies its values with (-1) if its score is greater or equal than half of the maximal score. For gene expression data, this amounts to treating under- and overexpression symmetrically. After the sign-flip procedure, low (expression) values point towards response class 0 and high (expression) values point towards class 1.

Usage

sign.flip(x, y)

Arguments

x

Numeric matrix of explanatory variables (\(p\) variables in columns, \(n\) cases in rows). For example, these can be microarray gene expression data which should be sign-flipped and then clustered.

y

Numeric vector of length \(n\) containing the class labels of the individuals. These labels have to be coded by 0 and 1.

Value

Returns a list containing:

flipped.matrix

The sign-flipped x-matrix.

signs

Numeric vector of length \(p\), which for each predictor variable indicates whether it was sign-flipped (coded by -1) or not (coded by +1).

See Also

wilma also for the references and score, as well as for a newer methodology, pelora and sign.change.

Examples

# NOT RUN {
data(leukemia, package="supclust")

op <- par(mfrow=c(1,3))
plot(leukemia.x[,69],leukemia.y)
title(paste("Margin = ", round(margin(leukemia.x[,69], leukemia.y),2)))

## Sign-flipping is very important
plot(leukemia.x[,161],leukemia.y)
title(paste("Margin = ", round(margin(leukemia.x[,161], leukemia.y),2)))
x <- sign.flip(leukemia.x, leukemia.y)$flipped.matrix
plot(x[,161],leukemia.y)
title(paste("Margin = ", round(margin(x[,161], leukemia.y),2)))
par(op)# reset
# }