QC_PCA_scoreplot: PCA score plot

In MWASTools: MWASTools: an integrated pipeline to perform metabolome-wide association studies

Description

This function generates a PCA score plot colored based on sample type (i.e. experimental or quality control (QC) sample). The plots generated with this function can be used to assess analytical reproducibility and stability. If the dataset is reproducible, all quality control samples should appear clustered in the center of the Hotelling's ellipse.

Usage

QC_PCA_scoreplot (PCA_model, metabo_SE, plot_labels = FALSE,  px = 1, py = 2,
                  CI_level = 0.95, pch = 20, xlim = NULL, ylim = NULL,
                  color_scale = c("cornflowerblue", "red"), grid = TRUE,...)

Arguments

PCA_model	"prcomp" object generated by the function "QC_PCA()".
metabo_SE	SummarizedExperiment object. See "MWAS_SummarizedExperiment()".
plot_labels	logical constant indicating whether the sample IDs will be displayed in the score plot.
px	numeric value indicating the index of the principal component that will be displayed on the x-axis.
py	numeric value indicating the index of the principal component that will be displayed on the y-axis.
CI_level	numeric value indicating the confidence interval for the Hotelling's ellipse.
pch	value specifying the symbol that will represent each sample in the score. To see all possible symbols, check "plot()" options.
xlim	numeric vector containing the minimum and maximum values of the x-axis.
ylim	numeric vector containing the minimum and maximum values of the y-axis.
color_scale	character vector corresponding to the 2-color scale that will be used to discriminate the experimental samples from the QC samples.
grid	logical constant indicating whether grid lines will be added to the plot.
...	other arguments passed to "plot()".

Value

A PCA score plot.

References

Fox J, Weisberg S. (2011). An R Companion to Applied Regression, Second Edition, Sage.

Mardia K, et al. (1979). Multivariate Analysis, London: Academic Press.

Examples

## Load data
data(metabo_SE)

## PCA model
PCA_model <- QC_PCA (metabo_SE)

## PCA score plots
QC_PCA_scoreplot (PCA_model, metabo_SE) # PC1 vs PC2
QC_PCA_scoreplot (PCA_model, metabo_SE, px = 3, py = 4) # PC3 vs PC4
QC_PCA_scoreplot(PCA_model, metabo_SE, plot_labels = TRUE) # show labels
QC_PCA_scoreplot (PCA_model, metabo_SE, CI_level = 0.80) # change CI

MWASTools documentation built on Nov. 8, 2020, 5:07 p.m.