mixomics_splsda_optimize: sPLS-DA

In antonvsdata/amp: Smoothing Metabolomics Analyses

Description

A wrapper for fitting an sPLS-DA model using splsda function of the mixOmics package. Automatically evaluates performance with different number of components and different number of features per component, then chooses the optimal number of components and optimal number of features for each component.

Usage

mixomics_splsda_optimize(object, y, ncomp_max, dist, n_features = c(1:10,
  seq(20, 300, 10)), ...)

Arguments

object	a MetaboSet object
y	character, column name of the grouping variable to predict
ncomp_max	numeric, the maximum number of components to try
dist	the distance metric to use, one of "max.dist", "mahalanobis.dist", "centroids.dist"
n_features	the number of features to try for each component
...	any parameters passed to mixOmics::plsda

Value

an object of class "splsdsa"

Examples

set.seed(38)
plsda_res <- mixomics_splsda_optimize(merged_sample, dist = "max.dist", y = "Group", ncomp_max = 2)

antonvsdata/amp documentation built on Jan. 8, 2020, 3:15 a.m.