Silke Szymczak and Cesaire J.K. Fouodo
This package provides different methods for identifying relevant variables in omics data sets using Random Forests. It implements the following approaches: empirical and parametric permutation (Altmann), Boruta, Vita, r2VIM (recurrent relative veriable importance), RFE (recursive feature elimination) and Hybrid, combining Vita and Boruta. All approaches use unscaled permutation variable importance and the R package ranger to generate the forests. The package also includes a function to simulate correlated gene expression data.
Installation from Github:
devtools::install_github("silkeszy/Pomona")
CRAN release coming soon.
For usage in R, see ?Pomona in R. Most importantly, see the Examples section. As a first example you could try
data <- simulation.data.cor(no.samples = 100, group.size = rep(10, 6), no.var.total = 200)
res <- var.sel.hybrid(x = data[, -1], y = data[, 1])
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.