PomaRandForest: Classification Random Forest for Mass Spectrometry Data

In POMA: User-friendly Workflow for Pre-processing and Statistical Analysis of Mass Spectrometry Data

Description

PomaRandForest() allows users to perform a classification Random Forest with a MS data matrix using the classical randomForest R package.

Usage

PomaRandForest(
  data,
  ntest = 20,
  ntree = 500,
  mtry = floor(sqrt(ncol(t(Biobase::exprs(data))))),
  nodesize = 1,
  nvar = 20
)

Arguments

data	A MSnSet object. First pData column must be the subject group/type.
ntest	Numeric indicating the percentage of observations that will be used as test set. Default is 20% of observations.
ntree	Number of trees to grow.
mtry	Number of variables randomly sampled as candidates at each split. This value is set sqrt(p) (where p is number of variables in data) by default.
nodesize	Minimum size of terminal nodes. By default is equal to 1.
nvar	Number of variables to show in the Gini plot.

Value

A list with all results for Random Forest including plots and data frames.

Author(s)

Pol Castellano-Escuder

References

A. Liaw and M. Wiener (2002). Classification and Regression by randomForest. R News 2(3), 18–22.

Examples

data("st000336")

st000336 %>% 
  PomaImpute() %>%
  PomaRandForest()

POMA documentation built on Nov. 8, 2020, 6:26 p.m.