plotIndiv: Plot of Individuals (Experimental Units)

In mixOmics: Omics Data Integration Project

Description

This function provides scatter plots for individuals (experimental units) representation in (sparse)(I)PCA, (regularized)CCA, (sparse)PLS(DA) and (sparse)(R)GCCA(DA).

Usage

## S3 method for class 'pls'
plotIndiv(object, comp = NULL, rep.space = NULL, ind.names = TRUE, group, col.per.group,
style = "ggplot2", ellipse = FALSE, ellipse.level = 0.95, centroid = FALSE, star = FALSE,
title = NULL, subtitle, legend = FALSE,
X.label = NULL, Y.label = NULL, Z.label = NULL, abline = FALSE,
xlim = NULL, ylim = NULL, col, cex, pch, pch.levels, alpha = 0.2, axes.box = "box",
layout = NULL,
size.title = rel(2), size.subtitle = rel(1.5), size.xlabel = rel(1),
size.ylabel = rel(1), size.axis = rel(0.8), size.legend = rel(1),
size.legend.title = rel(1.1), legend.title = "Legend", legend.title.pch = "Legend",
legend.position = "right",
point.lwd = 1, background = NULL, ... )

## S3 method for class 'mint.spls'
plotIndiv(object, comp = NULL, study = "global", rep.space = NULL, group, col.per.group,
style = "ggplot2", ellipse = FALSE, ellipse.level = 0.95, centroid = FALSE, star = FALSE,
title = NULL, subtitle, legend=FALSE,
X.label = NULL, Y.label = NULL, abline = FALSE,
xlim = NULL, ylim = NULL, col, cex, pch, layout = NULL,
size.title = rel(2), size.subtitle = rel(1.5), size.xlabel = rel(1),
size.ylabel = rel(1), size.axis = rel(0.8), size.legend = rel(1),
size.legend.title = rel(1.1), legend.title = "Legend", legend.position = "right",
point.lwd = 1, ... )

## S3 method for class 'sgcca'
plotIndiv(object, comp = NULL, blocks = NULL, ind.names = TRUE, group, col.per.group,
style = "ggplot2", ellipse = FALSE, ellipse.level = 0.95, centroid = FALSE, star = FALSE,
title = NULL, subtitle, legend = FALSE,
X.label = NULL, Y.label = NULL, Z.label = NULL, abline = FALSE,
xlim = NULL, ylim = NULL, col, cex, pch, pch.levels,
alpha = 0.2, axes.box = "box", layout = NULL,
size.title = rel(2), size.subtitle = rel(1.5), size.xlabel = rel(1),
size.ylabel = rel(1), size.axis = rel(0.8), size.legend = rel(1),
size.legend.title = rel(1.1), legend.title = "Legend", legend.title.pch = "Legend",
legend.position = "right",
point.lwd = 1, ... )

Arguments

object	object of class inheriting from any mixOmics: PLS, sPLS, PLS-DA, SPLS-DA, rCC, PCA, sPCA, IPCA, sIPCA, rGCCA, sGCCA, sGCCDA
comp	integer vector of length two (or three to 3d). The components that will be used on the horizontal and the vertical axis respectively to project the individuals.
rep.space	For objects of class "rcc", "pls", "spls", character string, (partially) matching one of "X-variate", "Y-variate" ,or "XY-variate", determining the subspace to project the individuals. Defaults to "X-variate" "pca" object and for "plsda" objects. For objects of class "pls" and "rcc",defaults, the tree subspaces represent the individuals. For objects of class "rgcca" and "sgcca", numerical value indicating the block data set form which to represent the individuals.
blocks	integer value of name of a block to be plotted using the GCCA module. See examples.
study	Indicates which study-specific outputs to plot. A character vector containing some levels of object$study, "all.partial" to plot all studies or "global" is expected. Default to "global".
ind.names	either a character vector of names for the individuals to be plotted, or FALSE for no names. If TRUE, the row names of the first (or second) data matrix is used as names (see Details).
group	factor indicating the group membership for each sample, useful for ellipse plots. Coded as default for the supervised methods PLS-DA, SPLS-DA,sGCCDA, but needs to be input for the unsupervised methods PCA, sPCA, IPCA, sIPCA, PLS, sPLS, rCC, rGCCA, sGCCA
col.per.group	character (or symbol) color to be used when 'group' is defined. Vector of the same length than the number of groups.
style	argument to be set to either 'graphics', 'lattice', 'ggplot2' or '3d' for a style of plotting. Default set to 'ggplot2'. See details. 3d is not available for MINT objects.
ellipse	boolean indicating if ellipse confidence region plots should be plotted. In the non supervised objects PCA, sPCA, IPCA, sIPCA, PLS, sPLS, rCC, rGCCA, sGCCA ellipse plot is only be plotted if the argument group is provided. In the PLS-DA, SPLS-DA,sGCCDA supervised object, by default the ellipse will be plotted accoding to the outcome Y.
ellipse.level	Numerical value indicating the confidence level of ellipse being plotted when ellipse =TRUE (i.e. the size of the ellipse). The default is set to 0.95, for a 95% region.
centroid	boolean indicating whether centroid points should be plotted. In the non supervised objects PCA, sPCA, IPCA, sIPCA, PLS, sPLS, rCC, rGCCA, sGCCA the centroid will only be plotted if the argument group is provided. The centroid will be calculated based on the group categories. In the supervised objects PLS-DA, SPLS-DA,sGCCDA the centroid will be calculated according to the outcome Y.
star	boolean indicating whether a star plot should be plotted, with arrows starting from the centroid (see argument centroid, and ending for each sample belonging to each group or outcome. In the non supervised objects PCA, sPCA, IPCA, sIPCA, PLS, sPLS, rCC, rGCCA, sGCCA star plot is only be plotted if the argument group is provided. In the supervised objects PLS-DA, SPLS-DA,sGCCDA the star plot is plotted according to the outcome Y.
title	set of characters indicating the title plot.
subtitle	subtitle for each plot, only used when several block or study are plotted.
legend	boolean. Whether the legend should be added. Default is FALSE.
X.label	x axis titles.
Y.label	y axis titles.
Z.label	z axis titles (when style = '3d').
abline	should the vertical and horizontal line through the center be plotted? Default set to FALSE
xlim,ylim	numeric list of vectors of length 2 and length =length(blocks), giving the x and y coordinates ranges.
col	character (or symbol) color to be used, possibly vector.
cex	numeric character (or symbol) expansion, possibly vector.
pch	plot character. A character string or a vector of single characters or integers. See points for all alternatives.
pch.levels	Only used when pch is different from col or col.per.group, ie when pch creates a second factor. Only used for the legend.
alpha	Semi-transparent colors (0 < 'alpha' < 1)
axes.box	for style '3d', argument to be set to either 'axes', 'box', 'bbox' or 'all', defining the shape of the box.
layout	layout parameter passed to mfrow. Only used when study is not "global"
size.title	size of the title
size.subtitle	size of the subtitle
size.xlabel	size of xlabel
size.ylabel	size of ylabel
size.axis	size of the axis
size.legend	size of the legend
size.legend.title	size of the legend title
legend.title	title of the legend
legend.title.pch	title of the second legend created by pch, if any.
legend.position	position of the legend, one of "bottom", "left", "top" and "right".
point.lwd	lwd of the points, used when ind.names = FALSE
background	color the background by the predicted class, see background.predict
...	external arguments or type par can be added with style = 'graphics'

Details

plotIndiv method makes scatter plot for individuals representation depending on the subspace of projection. Each point corresponds to an individual.

If ind.names=TRUE and row names is NULL, then ind.names=1:n, where n is the number of individuals. Also, if pch is an input, then ind.names is set to FALSE as we do not show both names and shapes.

plotIndiv can have a two layers legend. This is especially convenient when you have two grouping factors, such as a gender effect and a study effect, and you want to highlight both simulatenously on the graphical output. A first layer is coded by the group factor, the second by the pch argument. When pch is missing, a single layer legend is shown. If the group factor is missing, the col argument is used to create the grouping factor group. When a second grouping factor is needed and added via pch, pch needs to be a vector of length the number of samples. In the case where pch is a vector or length the number of groups, then we consider that the user wants a different pch for each level of group. This leads to a single layer legend and we merge col and pch. In the similar case where pch is a single value, then this value is used to represent all samples. See examples below for object of class plsda and splsda.

In the specific case of a single 'omics supervised model (plsda, splsda), users can overlay prediction results to sample plots in order to visualise the prediction areas of each class, via the background input parameter. Note that this functionality is only available for models with less than 2 components as the surfaces obtained for higher order components cannot be projected onto a 2D representation in a meaningful way. For more details, see background.predict

For customized plots (i.e. adding points, text), use the style = 'graphics' (default is ggplot2).

Note: the ellipse options were borrowed from the ellipse, see ?ellipse for more details about how the confidence region is calculated.

Author(s)

Ignacio González, Benoit Gautier, Francois Bartolo, Florian Rohart

See Also

text, background.predict, points and http://mixOmics.org/graphics for more details.

Examples

## plot of individuals for objects of class 'rcc'
# ----------------------------------------------------
data(nutrimouse)
X <- nutrimouse$lipid
Y <- nutrimouse$gene
nutri.res <- rcc(X, Y, ncomp = 3, lambda1 = 0.064, lambda2 = 0.008)

# default, only in the X space
plotIndiv(nutri.res)

## Not run: 
    # ellipse with respect to genotype in the XY space,
    # names also indicate genotype
    plotIndiv(nutri.res, rep.space= 'XY-variate',
    ellipse = TRUE, ellipse.level = 0.9,
    group = nutrimouse$genotype, ind.names = nutrimouse$genotype)

    # ellipse with respect to genotype in the XY space, with legend
    plotIndiv(nutri.res, rep.space= 'XY-variate', group = nutrimouse$genotype,
    legend = TRUE)


    # lattice style
    plotIndiv(nutri.res, rep.space= 'XY-variate', group = nutrimouse$genotype,
    legend = TRUE, style = 'lattice')

    # classic style, in the Y space
    plotIndiv(nutri.res, rep.space= 'Y-variate', group = nutrimouse$genotype,
    legend = TRUE, style = 'graphics')

## End(Not run)

## plot of individuals for objects of class 'pls' or 'spls'
# ----------------------------------------------------
data(liver.toxicity)
X <- liver.toxicity$gene
Y <- liver.toxicity$clinic
toxicity.spls <- spls(X, Y, ncomp = 3, keepX = c(50, 50, 50),
keepY = c(10, 10, 10))

#default
plotIndiv(toxicity.spls)

## Not run: 
# two layers legend: a first grouping with Time.Group and 'group'
# and a second with Dose.Group and 'pch'
plotIndiv(toxicity.spls, rep.space="X-variate", ind.name = FALSE,
group = liver.toxicity$treatment[, 'Time.Group'], # first factor
pch = as.numeric(factor(liver.toxicity$treatment$Dose.Group)), #second factor
pch.levels =liver.toxicity$treatment$Dose.Group, #levels of the second factor, for the legend
legend = TRUE)

## End(Not run)


# indicating the centroid
plotIndiv(toxicity.spls, rep.space= 'X-variate', ind.names = FALSE,
          group = liver.toxicity$treatment[, 'Time.Group'], centroid = TRUE)

# indicating the star and centroid
plotIndiv(toxicity.spls, rep.space= 'X-variate', ind.names = FALSE,
          group = liver.toxicity$treatment[, 'Time.Group'], centroid = TRUE, star = TRUE)


# indicating the star and ellipse
plotIndiv(toxicity.spls, rep.space= 'X-variate', ind.names = FALSE,
          group = liver.toxicity$treatment[, 'Time.Group'], centroid = TRUE,
            star = TRUE, ellipse = TRUE)



    # in the Y space, colors indicate time of necropsy, text is the dose
    plotIndiv(toxicity.spls, rep.space= 'Y-variate',
    group = liver.toxicity$treatment[, 'Time.Group'],
    ind.names = liver.toxicity$treatment[, 'Dose.Group'],
    legend = TRUE)


## plot of individuals for objects of class 'plsda' or 'splsda'
# ----------------------------------------------------
data(breast.tumors)
X <- breast.tumors$gene.exp
Y <- breast.tumors$sample$treatment

splsda.breast <- splsda(X, Y,keepX=c(10,10),ncomp=2)

# default option: note the outcome color is included by default!
plotIndiv(splsda.breast)

# also check ?background.predict for to visualise the prediction
# area with a plsda or splsda object!


## Not run: 
    # default option with no ind name: pch and color are set automatically
    plotIndiv(splsda.breast, ind.names = FALSE, comp = c(1, 2))

    # default option with no ind name: pch and color are set automatically, with legend
    plotIndiv(splsda.breast, ind.names = FALSE, comp = c(1, 2), legend = TRUE)

    # trying the different styles
    plotIndiv(splsda.breast, ind.names = TRUE, comp = c(1, 2),
    ellipse = TRUE, style = "ggplot2", cex = c(1, 1))
    plotIndiv(splsda.breast, ind.names = TRUE, comp = c(1, 2),
    ellipse = TRUE, style = "lattice", cex = c(1, 1))

    # changing pch of the two groups
    plotIndiv(splsda.breast, ind.names = FALSE, comp = c(1, 2),
    pch = c(15,16), legend = TRUE)

    # creating a second grouping factor with a pch of length 3,
    #  which is recycled to obtain a vector of length n
    plotIndiv(splsda.breast, ind.names = FALSE, comp = c(1, 2),
    pch = c(15,16,17), legend = TRUE)

    #same thing as
    pch.indiv = c(rep(15:17,15), 15, 16) # length n
    plotIndiv(splsda.breast, ind.names = FALSE, comp = c(1, 2),
    pch = pch.indiv, legend = TRUE)

    # change the names of the second legend with pch.levels
    plotIndiv(splsda.breast, ind.names = FALSE, comp = c(1, 2),
    pch = 15:17, pch.levels = c("a","b","c"),legend = TRUE)


## End(Not run)


## plot of individuals for objects of class 'mint.plsda' or 'mint.splsda'
# ----------------------------------------------------
data(stemcells)
res = mint.splsda(X = stemcells$gene, Y = stemcells$celltype, ncomp = 2, keepX = c(10, 5),
study = stemcells$study)

plotIndiv(res)

## Not run: 
#plot study-specific outputs for all studies
plotIndiv(res, study = "all.partial")

#plot study-specific outputs for study "2"
plotIndiv(res, study = "2")

## End(Not run)

## variable representation for objects of class 'sgcca' (or 'rgcca')
# ----------------------------------------------------
## Not run: 
data(nutrimouse)
Y = unmap(nutrimouse$diet)
data = list(gene = nutrimouse$gene, lipid = nutrimouse$lipid, Y = Y)
design1 = matrix(c(0,1,1,1,0,1,1,1,0), ncol = 3, nrow = 3, byrow = TRUE)
nutrimouse.sgcca <- wrapper.sgcca(X = data,
design = design1,
penalty = c(0.3, 0.5, 1),
ncomp = 3,
scheme = "horst")

# default style: one panel for each block
plotIndiv(nutrimouse.sgcca)

    # for the block 'lipid' with ellipse plots and legend, different styles
    plotIndiv(nutrimouse.sgcca, group = nutrimouse$diet, legend =TRUE,
    ellipse = TRUE, ellipse.level = 0.5, blocks = "lipid", title = 'my plot')
    plotIndiv(nutrimouse.sgcca, style = "lattice", group = nutrimouse$diet,
    legend = TRUE, ellipse = TRUE, ellipse.level = 0.5, blocks = "lipid",
    title = 'my plot')
    plotIndiv(nutrimouse.sgcca, style = "graphics", group = nutrimouse$diet,
    legend = TRUE, ellipse = TRUE, ellipse.level = 0.5, blocks = "lipid",
    title = 'my plot')

## End(Not run)

## variable representation for objects of class 'sgccda'
# ----------------------------------------------------
## Not run: 
# Note: the code differs from above as we use a 'supervised' GCCA analysis
data(nutrimouse)
Y = nutrimouse$diet
data = list(gene = nutrimouse$gene, lipid = nutrimouse$lipid)
design1 = matrix(c(0,1,0,1), ncol = 2, nrow = 2, byrow = TRUE)

nutrimouse.sgccda1 <- wrapper.sgccda(X = data,
Y = Y,
design = design1,
ncomp = 2,
keepX = list(gene = c(10,10), lipid = c(15,15)),
scheme = "centroid")


# plotIndiv
# ----------

# displaying all blocks. bu default colors correspond to outcome Y
plotIndiv(nutrimouse.sgccda1)


    # displaying only 2 blocks
    plotIndiv(nutrimouse.sgccda1, blocks = c(1,2), group = nutrimouse$diet)

    # with some ellipse, legend and title
    plotIndiv(nutrimouse.sgccda1, blocks = c(1,2), group = nutrimouse$diet,
    ellipse = TRUE, legend = TRUE, title = 'my sample plot')

## End(Not run)

Example output

Loading required package: MASS
Loading required package: lattice
Loading required package: ggplot2

Loaded mixOmics 6.14.0
Thank you for using mixOmics!
Tutorials: http://mixomics.org
Bookdown vignette: https://mixomicsteam.github.io/Bookdown
Questions, issues: Follow the prompts at http://mixomics.org/contact-us
Cite us:  citation('mixOmics')

Design matrix has changed to include Y; each block will be
            linked to Y.

mixOmics documentation built on June 1, 2018, 5:06 p.m.