seriation: pimage – R documentation

Become an expert in R — Interactive courses, Cheat Sheets, certificates and more!

pimage

Permutation Image Plot

Description

Provides methods for plotting image plots for matrix and dist objects given a permutation. By default, no permutation is performed. This plot can also be used as a more versatile replacement of image plot in graphics based on grid.

Usage

pimage(x, order = NULL, col = NULL, main ="", xlab = "", ylab = "",
  axes = "auto", zlim=NULL, key=TRUE, key.lab="", symkey=TRUE,
  upper.tri = TRUE, lower.tri = TRUE, prop = NULL,
  ..., newpage=TRUE, pop=TRUE, gp=NULL)

Arguments

`x`	a matrix or an object of class `dist`.
`order`	an object of class `ser_permutation`. If `NULL` the order in `x` is plotted.
`col`	a list of colors used. If `NULL`, a gray scale is used (for matrix larger values are displayed darker and for `dist` smaller distances are darker). For matrices containing logical data, black and white is used. For matrices containing negative values a symmetric diverging color palette is used.
`main`	plot title.
`xlab, ylab`	labels for the x and y axes.
`axes`	a character string indicating if axes labels (column and row names of `x`) should be potted. Possible values are `"auto"` (only plot if less then 25 labels), `"x"`, `"y"`, `"both"` and `"none"`.
`zlim`	vector with two elements giving the range (min, max) for representing the values in the matrix.
`key`	logical; add a color key? No key is available for logical matrices.
`key.lab`	string plotted next to the color key.
`symkey`	logical; if `x` contains negative values, should the color palate be symmetric (zero is in the middle)>
`upper.tri, lower.tri`	a logical indicating whether to show the upper or lower triangle of the distance matrix.
`prop`	logical; draw the cells in the image proportional (defaults to `TRUE` for `dist` and `FALSE` for matrix).
`...`	further arguments passed on to `image` in graphics.
`newpage, pop`	two logical. Start plot on a new page and pop the viewports after plotting (see grid).
`gp`	a `gpar` object (see grid).

Details

Plots a matrix in its original row and column orientation. This means, in a plot the columns become the x-coordinates and the reversed rows the y-coordinates.

If x is of class dist it is converted to full-storage representation before plotting.

The viewports used for plotting are called: "plot", "image" and "colorkey".

Note: Since pimage uses grid, it should not be mixed with base R primitive plotting functions, but the appropriate functions in grid-package.

Author(s)

Christian Buchta and Michael Hahsler

Examples

x <- matrix(sample(c(FALSE, TRUE), 300, rep=TRUE), ncol=10,
  dimnames = list(1:30, LETTERS[1:10]))

## matrix (large values are dark/black)
pimage(x, main = "Random data", key = FALSE)

## plot seriated matrix (use red, proportional display and plot all axes)
pimage(x, seriate(x), col = c("white", "red"),
  prop = TRUE, axes="both", main = "Reordered data", key = FALSE)

## show correlation (for neg. values a diverging color scheme is
## used automatically)
pimage(cor(x), prop=TRUE)

## distances (note that low distances are represented dark!)
d <- dist(x, method = "binary")
pimage(d, upper.tri = FALSE, main = "Distances")

pimage(d, seriate(d), upper.tri = FALSE, main = "Distances", axes = "both")

## add to the plot using functions in package grid
library("grid")
pimage(x, pop = FALSE)           ### pop = FALSE allows us to manipulate viewports
downViewport(name = "image")

## highlight cell 7/5 with a red arrow
grid.lines(x = c(5, 7), y = c(3, 5), arrow = arrow(),
  default.units = "native", gp = gpar(col="red", lwd = 3))

## add a red box around rows 15 and 16
grid.rect(x = 0.5, y = 15.5, width = ncol(x), height = 2,
  just = "left",
  default.units = "native", gp = gpar(col="red", lwd = 3, fill = NA))

## remove the viewports
popViewport(0)

## put several pimages on a page (uses viewports and newpage = FALSE)
library(grid)
grid.newpage()
pushViewport(viewport(layout=grid.layout(nrow = 1, ncol = 2)))
pushViewport(viewport(layout.pos.row = 1, layout.pos.col = 1))

## seriate matrix
o <- seriate(x)
pimage(x, o, main = "Random data", prop = TRUE, axes = "both", key = FALSE,
  newpage = FALSE)

upViewport(1)
pushViewport(viewport(layout.pos.row = 1, layout.pos.col = 2))

## add the reordered disimilarity matrix for rowa
pimage(d, o[[1]], upper.tri = FALSE, main = "Distances", axes = "both",
  key = FALSE, newpage = FALSE)

upViewport(1)
popViewport(0)

seriation

Infrastructure for Ordering Objects Using Seriation

v1.2-9

GPL-3

Authors

Michael Hahsler [aut, cre, cph], Christian Buchta [aut, cph], Kurt Hornik [aut, cph], Fionn Murtagh [ctb, cph], Michael Brusco [ctb, cph], Stephanie Stahl [ctb, cph], Hans-Friedrich Koehn [ctb, cph]

Initial release

2020-09-29