The loadingplot function represents positive values of a vector
and identifies the values above a given threshold. It can also
indicate groups of observations provided as a factor.
Such graphics can be used, for instance, to assess the weight of each variable (loadings) in a given analysis.
loadingplot(x, ...)# S3 method for default
loadingplot(x, at=NULL, threshold=quantile(x,0.75),
axis=1, fac=NULL, byfac=FALSE,
lab=NULL, cex.lab=0.7, cex.fac=1, lab.jitter=0,
main="Loading plot", xlab="Variables", ylab="Loadings",
srt = 0, adj = NULL, ...)
Invisibly returns a list with the following components:
- threshold: the threshold used
- var.names: the names of observations above the threshold
- var.idx: the indices of observations above the threshold
- var.values: the values above the threshold
either a vector with numeric values to be plotted, or a
matrix-like object containing numeric values. In such case, the
x[,axis] is used as vector of values to be plotted.
an optional numeric vector giving the abscissa at which loadings are plotted. Useful when variates are SNPs with a known position in an alignement.
a threshold value above which values of x are identified. By default, this is the third quartile of x.
an integer indicating the column of x to be plotted; used only if x is a matrix-like object.
a factor defining groups of observations.
a logical stating whether loadings should be averaged by
groups of observations, as defined by fac.
a character vector giving the labels used to annotate values above the threshold; if NULL, names are taken from the object.
a numeric value indicating the size of annotations.
a numeric value indicating the size of annotations for groups of observations.
a numeric value indicating the factor of randomisation for the position of annotations. Set to 0 (by default) implies no randomisation.
the main title of the figure.
the title of the x axis.
the title of the y axis.
rotation of the labels; see ?text.
adjustment of the labels; see ?text.
further arguments to be passed to the plot function.
Thibaut Jombart t.jombart@imperial.ac.uk
x <- runif(20)
names(x) <- letters[1:20]
grp <- factor(paste("group", rep(1:4,each=5)))
## basic plot
loadingplot(x)
## adding groups
loadingplot(x,fac=grp,main="My title",cex.lab=1)
Run the code above in your browser using DataLab