Skip to contents

Plot sample distribution

Source: R/analysis.R
plotDistribution.Rd

The tooltip shows the median, variance, maximum, minimum and number of non-NA samples of each data series, as well as sample names if available.

Usage

plotDistribution(
  data,
  groups = NULL,
  rug = length(data) < 500,
  vLine = TRUE,
  ...,
  title = NULL,
  subtitle = NULL,
  type = c("density", "boxplot", "violin"),
  invertAxes = FALSE,
  psi = NULL,
  rugLabels = FALSE,
  rugLabelsRotation = 0,
  legend = TRUE,
  valueLabel = NULL
)

Arguments

data

Numeric, data frame or matrix: gene expression data or alternative splicing event quantification values (sample names are based on their names or colnames)

groups

List of sample names or vector containing the group name per data value (read Details); if NULL or a character vector of length 1, data values are considered from the same group

rug

Boolean: show rug plot?

vLine

Boolean: plot vertical lines (including descriptive statistics for each group)?

...

Arguments passed on to stats::density.default

bw

the smoothing bandwidth to be used. The kernels are scaled such that this is the standard deviation of the smoothing kernel. (Note this differs from the reference books cited below, and from S-PLUS.)

bw can also be a character string giving a rule to choose the bandwidth. See bw.nrd.
The default, "nrd0", has remained the default for historical and compatibility reasons, rather than as a general recommendation, where e.g., "SJ" would rather fit, see also Venables and Ripley (2002).

The specified (or computed) value of bw is multiplied by adjust.

adjust

the bandwidth used is actually adjust*bw. This makes it easy to specify values like ‘half the default’ bandwidth.

kernel,window

a character string giving the smoothing kernel to be used. This must partially match one of "gaussian", "rectangular", "triangular", "epanechnikov", "biweight", "cosine" or "optcosine", with default "gaussian", and may be abbreviated to a unique prefix (single letter).

"cosine" is smoother than "optcosine", which is the usual ‘cosine’ kernel in the literature and almost MSE-efficient. However, "cosine" is the version used by S.

weights

numeric vector of non-negative observation weights, hence of same length as x. The default NULL is equivalent to weights = rep(1/nx, nx) where nx is the length of (the finite entries of) x[]. If na.rm = TRUE and there are NA's in x, they and the corresponding weights are removed before computations. In that case, when the original weights have summed to one, they are re-scaled to keep doing so.

Note that weights are not taken into account for automatic bandwidth rules, i.e., when bw is a string. When the weights are proportional to true counts cn, density(x = rep(x, cn)) may be used instead of weights.

width

this exists for compatibility with S; if given, and bw is not, will set bw to width if this is a character string, or to a kernel-dependent multiple of width if this is numeric.

give.Rkern

logical; if true, no density is estimated, and the ‘canonical bandwidth’ of the chosen kernel is returned instead.

subdensity

used only when weights are specified which do not sum to one. When true, it indicates that a “sub-density” is desired and no warning should be signalled. By default, when false, a warning is signalled when the weights do not sum to one.

warnWbw

logical, used only when weights are specified and bw is character, i.e., automatic bandwidth selection is chosen (as by default). When true (as by default), a warning is signalled to alert the user that automatic bandwidth selection will not take the weights into account and hence may be suboptimal.

n

the number of equally spaced points at which the density is to be estimated. When n > 512, it is rounded up to a power of 2 during the calculations (as fft is used) and the final result is interpolated by approx. So it almost always makes sense to specify n as a power of two.

from,to

the left and right-most points of the grid at which the density is to be estimated; the defaults are cut * bw outside of range(x).

cut

by default, the values of from and to are cut bandwidths beyond the extremes of the data. This allows the estimated density to drop to approximately zero at the extremes.

title

Character: plot title

subtitle

Character: plot subtitle

type

Character: density, boxplot or violin plot

invertAxes

Boolean: plot X axis as Y and vice-versa?

psi

Boolean: are data composed of PSI values? If NULL, psi = TRUE if all data values are between 0 and 1

rugLabels

Boolean: plot sample names in the rug?

rugLabelsRotation

Numeric: rotation (in degrees) of rug labels; this may present issues at different zoom levels and depending on the proximity of data values

legend

Boolean: show legend?

valueLabel

Character: label for the value (by default, either Inclusion levels or Gene expression)

Value

highchart object with density plot

Details

Argument groups can be either:

  • a list of sample names, e.g. list("Group 1"=c("Sample A", "Sample B"), "Group 2"=c("Sample C")))

  • a character vector with the same length as data, e.g. c("Sample A", "Sample C", "Sample B").

See also

Other functions to perform and plot differential analyses: diffAnalyses()

Examples

data   <- sample(20, rep=TRUE)/20
groups <- paste("Group", c(rep("A", 10), rep("B", 10)))
names(data) <- paste("Sample", seq(data))
plotDistribution(data, groups)



# Using colours
attr(groups, "Colour") <- c("Group A"="pink", "Group B"="orange")
plotDistribution(data, groups)