MAVTp: Random Forests-based procedure

In MAVTgsa: Three methods to identify differentially expressed gene sets, ordinary least square test, Multivariate Analysis Of Variance test with n contrasts and Random forest.

Description

A Random Forests-based procedure is to identify gene sets that can accurately predict samples from different experimental conditions or are associated with the continuous phenotypes.

Usage

MAVTp(DATA, GS, nbPerm = 5000, numoftree = 500, type = c("cont", "cate"), impt = TRUE)

Arguments

DATA	a gene expression data matrix with samples in columns. The first row contains the information of the experimental condition of each sample. The remaining rows contain gene expression.
GS	an m x k binary matrix with code (0, 1), where k is the number of gene sets. Each column represents a pre-defined gene set.
nbPerm	the number of permutation specified
numoftree	the number of trees to grow
type	This can be one of "cont" (continuous phenotypes) and "cate" (categorical phenotypes).
impt	If TRUE (default), the importance measurement will be output.

Value

A list of the p-values of random forests for GSA. The importance measurement of individual genes for those significant gene sets will also be output when impt is set TRUE.

Note

R > 2.14.0

Author(s)

Chih-Yi Chien, Chen-An Tsai, Ching-Wei Chang, and James J. Chen

References

H.M. Hsueh, et al. (2013) Random forests-based differential analysis of gene sets for gene expression data. Gene, 518, 179-186.

Examples

data(data)
data(GS)
a=proc.time()
MAVTp(data,GS, nbPerm = 50, numoftree = 500, type = "cate", impt = TRUE)
proc.time()-a

MAVTgsa documentation built on May 2, 2019, 6:37 a.m.