N-Way Partial Least Squares Modelling of Multi-Way Data.

NPLStoolbox

Overview

The NPLStoolbox allows researchers to use the N-way Partial Least Squares method for their multi-way data.

• ncrossreg() allows the user to identify the appropriate number of NPLS components for their data.
• triPLS1() allows the user to create an NPLS model.
• npred() allows the user to predict y for new data.

This package also comes with an example dataset Cornejo2025: a clinical observational cohort study of 39 transgender persons starting gender-affirming hormone therapy, containing longitudinally measured tongue microbiome, salivary microbiome, salivary cytokine, salivary biochemistry, and circulatory hormone levels (doi TBD).

Documentation

A basic introduction to the package using the example dataset is given in vignette("Cornejo2025_analysis").

This vignette and all function documentation can be found here.

Installation

The NPLStoolbox package can be installed from CRAN using:

install.packages("NPLStoolbox")

Development version

You can install the development version of NPLStoolbox from GitHub with:

# install.packages("pak")
pak::pak("GRvanderPloeg/NPLStoolbox")

Usage

library(parafac4microbiome)
library(NPLStoolbox)
set.seed(123)

# Process one of the data cubes from Cornejo2025
processedTongue = processDataCube(Cornejo2025$Tongue_microbiome, sparsityThreshold=0.5, considerGroups=TRUE, groupVariable="GenderID", centerMode=1, scaleMode=2)

# Prepare Y: binarized gender identity
Y = as.numeric(as.factor(Cornejo2025$Tongue_microbiome$mode1$GenderID))
Ycnt = Y - mean(Y)

# Make a one-component NPLS model
model = triPLS1(processedTongue$data, Ycnt, 1)

Getting help

If you encounter an unexpected error or a clear bug, please file an issue with a minimal reproducible example here on Github. For questions or other types of feedback, feel free to send an email.