plot.combi: Make multiplots of the data integration object

In CenterForStatistics-UGent/compIntegrate: Compositional omics model based visual integration

Make multiplots of the data integration object

Description

Make multiplots of the data integration object

Usage

## S3 method for class 'combi'
plot(
  x,
  ...,
  Dim = c(1, 2),
  samDf = NULL,
  samShape = NULL,
  samCol = NULL,
  featurePlot = "threshold",
  featNum = 15L,
  samColValues = NULL,
  manExpFactorTaxa = 0.975,
  featSize = switch(featurePlot, threshold = 2.5, points = samSize * 0.7, density = 0.35),
  crossSize = 4,
  manExpFactorVar = 0.975,
  varNum = nrow(x$alphas),
  varSize = 2.5,
  samSize = 1.75,
  featCols = c("darkblue", "darkgreen", "grey10", "turquoise4", "blue", "green", "grey",
    "cornflowerblue", "lightgreen", "grey75"),
  strokeSize = 0.05,
  warnMonotonicity = FALSE,
  returnCoords = FALSE,
  squarePlot = TRUE,
  featAlpha = 0.5,
  featShape = 8,
  xInd = 0,
  yInd = 0,
  checkOverlap = FALSE,
  shapeValues = (21:(21 + length(unique(samDf[[samShape]]))))
)

Arguments

x	the fit
...	additional arguments, currently ignored
Dim	the dimensions to be plotted
samDf	a dataframe of sample variables
samShape	A variable name from samDf used to shape the samples
samCol	A variable name from samDf used to colour the samples
featurePlot	A character string, either "threshold", "points" or "density". See details
featNum, varNum	The number of features and variables to plot
samColValues	Colours for the samples
manExpFactorTaxa, manExpFactorVar	Expansion factors for taxa and variables, normally calculated natively
featSize, crossSize, varSize, samSize, strokeSize	Size parameters for the features (text, dots or density contour lines), central cross, variable labels, sample dots, sample strokes and feature contour lines
featCols	Colours for the features
warnMonotonicity	A boolean, should a warning be thrown when the feature proportions of compositional views do not all vary monotonically with all latent variables?
returnCoords	A boolean, should coordinates be returned, e.g. for use in thrird party software
squarePlot	A boolean, should the axes be square? Strongly recommended
featAlpha	Controls the transparacny of the features
featShape	Shape of feature dots when featurePlot = "points"
xInd, yInd	x and y indentations
checkOverlap	A boolean, should overlapping labels be omitted?
shapeValues	the shapes, as numeric values

Details

It is usually impossible to plot all features with their labels. Therefore, he default option of the 'featurePlot' parameter is "threshold", whereby only the 'featNum" features furthest away from the origin are shown. Alternatively, the "points" or "density" options are available to plot all features as a point or density cloud, but without labels.

Value

A ggplot object containing the plot

Examples

data(Zhang)
## Not run: 
#Unconstrained
microMetaboInt = combi(
list("microbiome" = zhangMicrobio, "metabolomics" = zhangMetabo),
distributions = c("quasi", "gaussian"), compositional = c(TRUE, FALSE),
logTransformGaussian = FALSE, verbose = TRUE)
#Constrained
microMetaboIntConstr = combi(
    list("microbiome" = zhangMicrobio, "metabolomics" = zhangMetabo),
    distributions = c("quasi", "gaussian"), compositional = c(TRUE, FALSE),
    logTransformGaussian = FALSE, covariates = zhangMetavars, verbose = TRUE)
## End(Not run)
    #Load the fits
load(system.file("extdata", "zhangFits.RData", package = "combi"))
plot(microMetaboInt)
plot(microMetaboInt, samDf = zhangMetavars, samCol = "ABX")
#Plot all features as points or density
plot(microMetaboInt, samDf = zhangMetavars, samCol = "ABX",
featurePlot = "points")
plot(microMetaboInt, samDf = zhangMetavars, samCol = "ABX",
featurePlot = "density")
#Constrained
plot(microMetaboIntConstr, samDf = zhangMetavars, samCol = "ABX")

CenterForStatistics-UGent/compIntegrate documentation built on Aug. 4, 2023, 1:08 p.m.