randomForest {MetNet}R Documentation

Create a adjacency matrix based on random forest

Description

randomForest infers an adjacency matrix using random forest using the rfPermute function from the rfPermute package. randomForest extracts the p-values by the function rp.importance and writes the presence/absence based on the significance value (α ≤q 0.05) of this connection to a matrix. The adjacency matrix is returned.

Usage

randomForest(x, parallel=FALSE, randomForest_adjust="none", ...)

Arguments

x

matrix, where columns are the samples and the rows are features (metabolites), cell entries are intensity values

parallel

logical, should computation be parallelized? If parallel=TRUE the bplapply will be applied if parallel=FALSE the lapply function will be applied.

randomForest_adjust

character, correction method for p-values from rp.importance, randomForest_adjust will be passed to the p.adjust function and should be one of "holm", "hochberg", "hommel", "bonferroni", "BH", "BY", "fdr", "none"

...

parameters passed to rfPermute.default

Details

For use of the parameters used in the rfPermute function, refer to ?rfPermute::rfPermute.default.

Value

matrix, matrix with edges inferred from random forest algorithm rfPermute and rp.importance

Author(s)

Thomas Naake, thomasnaake@googlemail.com

Examples

data("x_test", package="MetNet")
x <- x_test[, 3:dim(x_test)[2]]
x <- as.matrix(x)
## Not run: randomForest(x)
[Package MetNet version 1.2.0 Index]