Theory, Examples & Exercises
Section: Ordination analysis
vegan) - without environmental variables, the function calculates PCoA, while with environmental variables it calculates distance-based RDA. Input could be either species composition matrix (samples x species) or distance matrix (in that case, the species scores will not be available, unless the original species composition matrix is provided as argument
comm). By default
distance = “euclidean”, which returns results identical to PCA. Note that even if no environmental variables are included, the formula structure is still required (e.g.
capscale (spe ~ 1, distance = 'bray')).
stats) - calculates PCoA on matrix of distances among samples (this could be calculated e.g. by function
vegan). Use function
ordiplotto project the ordination diagram.
vegan) - based on
cmdscalefunction, but allows to weight the importance of samples in the PCoA. If arguments
eig = TRUEor
x.ret = TRUE, the function returns an object of class “wcmdscale” with print, plot, scores, eigenvals and stressplot methods.
ape) - another way how to achieve PCoA analysis. Use
biplot.pcoafunction (or simply generic
biplot) to project ordination diagram. Does not work with
vegan) - rather advanced function, composed of many subroutine steps. See example below for details.
vegan) - draws Shepards stress plot, which is the relationship between real distances between samples in resulting m dimensional ordination solution, and their particular compositional dissimilarities expressed by selected dissimilarity measure.
vegan) - returns goodness-of-fit of particular samples. See example how can be this result visualized (inspired by Borcard et al. 2011).