monogeneaGM (version 1.1)

pcloadhm: Heat map representation of principal component loadings of shape variables

Description

This function produces a graphical representation of selected principal component loadings of shape variables in the form of a heat map.

Usage

pcloadhm(x, pc = 1, sgn = 1, nrow = 11, color.code = NULL, ylab = "Landmark", xlab = "Coordinate", yaxis = FALSE, tit = NULL)

Arguments

x

a matrix containing the loadings of shape variables (row) of each principal component (column)

pc

a constant specifying the principal component of interest

sgn

the sign of the loadings; this value should follow the one used in pca2d

nrow

a constant indicating the number of landmarks defined in the anchors; defaults to 11

color.code

a character vector of hex color codes that define the color palette; if left undefined, defaults to the red-white-blue palette

ylab

y-axis title for the plot

xlab

x-axis title for the plot

yaxis

if TRUE, the y-axis values are labelled

tit

title for the plot

Details

The sign and magnitude of loadings of shape variables for a particular principal component is important for the latter's biological interpretation. A heat map representation is an alternative to the usual manner of presenting them in tabular form, and may be more effective for presentation purpose. Reference to the circular plots (plotCircular) for each landmark and the PCA plots can be very useful in determining the biological interpretation of a particular principal component.

References

Khang TF, Soo OYM, Tan WB, Lim LHS. (2016). Monogenean anchor morphometry: systematic value, phylogenetic signal, and evolution. PeerJ 4:e1668.

See Also

plotCircular, pca2d, colorBar

Examples

library(phytools)

data(ligophorus_shape)
data(ligotree)
data(spcolmap)

shapev <- pca2d(ligophorus_shape[,1:22], sgn=1, labcol=spcolmap$color,
phylo=TRUE, phy=ligotree, genus="L. ",
bound.y = c(-0.1, 0.1), bound.x1 =c(-0.15,0.2), bound.x2=c(-0.15,0.2))

fff <- c(0,1,1,2,2,3,3,0,4,4)
nf <- layout(matrix(c(rep(0,length(fff)),rep(fff,5),rep(0,length(fff))),
7,length(fff),byrow=TRUE))
layout.show(nf)
par(mar=c(5,4,4,1))
#the loadings for the first three PC of shape variables of the ventral anchors
pcloadhm(shapev$variable,sgn=1,pc=1,yaxis=TRUE,tit="VPC1")
pcloadhm(shapev$variable,sgn=1,pc=2,ylab="", tit="VPC2")
pcloadhm(shapev$variable,sgn=1,pc=3,ylab="", tit="VPC3")

#add a colorbar for completeness
par(mar=c(5,2,4,3))
colorBar(min=-1, max=1)