plotArrow: Arrow sample plot

In mixOmicsTeam/mixOmics: Omics Data Integration Project

Arrow sample plot

Description

Represents samples from multiple coordinates to assess the alignment in the latent space.

Usage

plotArrow(
  object,
  comp = c(1, 2),
  ind.names = TRUE,
  group = NULL,
  col = NULL,
  ind.names.position = c("start", "end"),
  ind.names.size = 2,
  pch = NULL,
  pch.size = 2,
  arrow.alpha = 0.6,
  arrow.size = 0.5,
  arrow.length = 0.2,
  legend = if (is.null(group)) FALSE else TRUE,
  legend.title = NULL,
  ...
)

Arguments

object	object of class inheriting from mixOmics: PLS, sPLS, rCC, 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.
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). If 'pch' is set this will overwrite the names as shapes. Default is TRUE. Not avaliable for MINT objects.
group	Factor indicating the group membership for each sample.
col	character (or symbol) color to be used. If group provided, should be a vector of the same length as groups, order of colours will be respected to correspond to order of group levels.
ind.names.position	One of c('start', 'end') indicating where to show the ind.names . Not used in block analyses, where centroids are used.
ind.names.size	Numeric, sample name size.
pch	plot character. A character string or a named vector of single characters or integers whose names match those of object$variates.
pch.size	Numeric, sample point character size.
arrow.alpha	Numeric between 0 and 1 determining the opacity of arrows.
arrow.size	Numeric, variable arrow head size.
arrow.length	Numeric, length of the arrow head in 'cm'.
legend	Logical, whether to show the legend if group != NULL.
legend.title	Character, the legend title if group != NULL.
...	Not currently used. sample size to display sample names.

Details

Graphical of the samples (individuals) is displayed in a superimposed manner where each sample will be indicated using an arrow. The start of the arrow indicates the location of the sample in X in one plot, and the tip the location of the sample in Y in the other plot. Short arrows indicate a strong agreement between the matching data sets, long arrows a disagreement between the matching data sets. The representation space is scaled using the range of coordinates so minimum and maximum values are equal for all blocks. Since the algorithm maximises the covariance of these components, the absolute values do not affect the alignment.

For objects of class "GCCA" and if there are more than 2 blocks, the start of the arrow indicates the centroid between all data sets for a given individual and the tips of the arrows the location of that individual in each block.

Value

A ggplot object

Author(s)

Al J Abadi

References

Lê Cao, K.-A., Martin, P.G.P., Robert-Granie, C. and Besse, P. (2009). Sparse canonical methods for biological data integration: application to a cross-platform study. BMC Bioinformatics 10:34.

See Also

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

Examples

## 'spls' class - examples demonstrate how to control sample colours with sample names shown
# ------------------------------------------------------
data(liver.toxicity)
X <- liver.toxicity$gene
Y <- liver.toxicity$clinic
spls.obj <- spls(X, Y, ncomp = 3, keepX = c(50, 50, 50),
                 keepY = c(10, 10, 10))

# colors indicate time of necropsy, text is the dose, label at start of arrow
plotArrow(spls.obj,  group = as.factor(liver.toxicity$treatment[, 'Time.Group']),
          col = c("red", "blue", "purple", "darkgreen"), 
          ind.names  = liver.toxicity$treatment[, 'Dose.Group'],
          legend = TRUE, position.names = 'start', legend.title = 'Time.Group')


## 'rcc' class - examples demonstrate how to control shape of all samples
# -------------------------------------------------------------------------------

# create model
data(nutrimouse)
X <- nutrimouse$lipid
Y <- nutrimouse$gene
rcc.obj <- rcc(X, Y, ncomp = 3, lambda1 = 0.064, lambda2 = 0.008)

primary_groups <- nutrimouse$diet

# plot samples coloured by primary groups, by default shapes are all circles
plotArrow(rcc.obj, ind.names = FALSE,
          group = primary_groups, legend = TRUE)

# plot samples coloured by primary groups, force all samples to have 
# a different shape (2 = triangle)
plotArrow(rcc.obj, ind.names = FALSE,
          group = primary_groups, legend = TRUE,
          pch = 2)


## 'sgccda' class - examples demonstrate how to control shape of different blocks
# --------------------------------------------------------------------------------

data(breast.TCGA)
idx = seq(1, length(breast.TCGA$data.train$subtype), 10)
X <- list(mRNA = breast.TCGA$data.train$mrna[idx,],
          miRNA = breast.TCGA$data.train$mirna[idx,],
          protein = breast.TCGA$data.train$protein[idx,])
Y <- breast.TCGA$data.train$subtype[idx] # set the response variable

diablo.obj <- block.splsda(X, Y, ncomp = 2) # undergo multiblock sPLS-DA

# plot the samples using an arrow plot - 
plotArrow(diablo.obj, 
          ind.names = FALSE,
          legend = TRUE,
          title = 'TCGA, DIABLO comp 1 - 2') 

pchs <- c(3, 2, 1)
names(pchs) <- c("miRNA", "mRNA", "protein")

# plot the samples using an arrow plot changed shapes of data blocks
plotArrow(diablo.obj, 
          ind.names = FALSE,
          legend = TRUE,
          title = 'TCGA, DIABLO comp 1 - 2',
          pch = pchs) 

mixOmicsTeam/mixOmics documentation built on April 13, 2025, 11:59 p.m.