mds: mds class

In RFmarkerDetector: Multivariate Analysis of Metabolomics Data using Random Forests

Description

A constructor function for the S3 class mds; the mds class encapsulates useful information for generating MDS plots

Usage

mds(dataset, opt = list(ntree = 1000, mtry = round(sqrt(ncol(dataset) - 2)),
  seed = 1234))

Arguments

dataset	a n x p dataframe representing the dataset to be used to train the model. The first two columns must represent respectively the samples names and the class labels related to each sample
opt	a list of optional parameters useful to train the random forest. It may include the number of trees (ntree), the parameter mtry and the seed

Value

an object of class mds including the following attributes:

• model an object of class random forest containing the proximity matrix

• classes a factor with the classes associated with each sample, used to train the model

• sample_names the vector of the names of the samples

Author(s)

Piergiorgio Palla

Examples

data(cachexiaData)
params = list(ntree = 1000, mtry = round(sqrt(ncol(cachexiaData) -2)), seed = 1234)
mds_obj <- mds(cachexiaData, opt = params)

RFmarkerDetector documentation built on May 2, 2019, 3:42 p.m.