weight.mbmr: Computation of weights for informative gene selection using...

In BootMRMR: Bootstrap-MRMR Technique for Informative Gene Selection

Description

Weights associated with genes in a dataset computed from the Modified Bootstrap MRMR technique will provide a reliable measure for informative gene selection.

Usage

weight.mbmr(x, y, m, s, plot)

Arguments

x	x is a N by p dataframe of gene expression, where rows are genes and columns are as samples/subjects (gene names are taken as row names).
y	y is a p by 1 numeric vector with entries 1 and -1 representing sample labels, where 1 and -1 represents the sample label of subjects/ samples for stress and control condition respectively.
m	m is a scalar representing the size of the Modified Bootstrap Sample (i.e. Out of p samples/subjects, m samples/subjects are randomly drawn with replacement, which constitutes one Modified Bootstrap Sample).
s	s is a scalar representing the number of Modified Bootstrap samples (i.e. number of times each of the m samples/subjects will be resampled from p samples/subjects).
plot	plot is a character string must either take logical value TRUE/FALSE representing whether to plot the weights of genes in the dataset.

Details

The function returns a vector of weights associated with each genes in the dataset using Modified Bootstrap MRMR technique.

Author(s)

Samarendra Das

References

Wang J, Chen L, Wang Y, Zhang J, Liang Y, Xu D (2013) A Computational systems biology study for understanding salt tolerance mechanism in Rice. PLoS one 8(6): e64929.

Examples

data(rice_salt)
x=as.data.frame(rice_salt[-1,])
y=as.numeric(rice_salt[1,])
m=36
s=80
weight.mbmr(x, y, m, s, plot=FALSE)

BootMRMR documentation built on May 1, 2019, 7:49 p.m.