The phylogenetic eigenvector regression (PVR) starts by performing an eigendecomposition of a pairwise double-centered phylogenetic distance matrix between species. The eigenvectors (representing the traits under analysis) estimated values express phylogenetic trends in data and residuals express independent evolution of each species.
PVR(x,phy,trait,envVar,method="moran",weights,
scaled=FALSE,sig=TRUE,sig.t=0.05,MI.t=0.05,psr.t=0.01,
accvalue.t=0.9,...)An object of class PVR (created by the PVRdecomp function) or class PSR (requiered by the "PSR" method).
An object of class phylo that contains an ultrametric phylogeny.
A vector, data frame or matrix that contains traits sets (for data frames and matrices, each column must represent a trait set).
A vector, data frame or matrix that contains environmental variables. Used to estimates the variation of a trait set that is explained by phylogeny and by environment.
Character string. A name for the eigenvectors selection method. It can be "moran", "stepwise", "psr" or "sequential".
Weighting matrix based on Phylogenetic distances used in the "moran" method. If no weights matrix is provided, weights will be set to max(D) - Dij, where D is the phylogenetic distance matrix.
Logical. Should the phylogenetic distances be scaled into the range of 0 to 1. Default is FALSE.
Logical. Should the eigenvectors selected by the "moran" method be selected by the significance of residuals autocorrelation. If FALSE the eigenvectors will be selected by Moran's I values.
Minimum residuals Moran's I value used to select eigenvectors when significance is FALSE.
The significance treshold used to select eigenvectors by the "moran" method.
The minimum acumulate R2 gain treshold used to select eigenvectors by the "PSR" method.
Relative accumulated eigenvalue treshold use to select the eigenvectors by the "sequential" method.
Parameters passed to the stepwise regression used in the "AIC" method
A PVR class object.
Diniz-Filho, J.A.F., Sant'Ana, C.E.R. and Bini, L.M. (1998). An eigenvector method for estimating phylogenetic inertia. Evolution, 52:1247-1262.
Legendre, P. and Legendre, L. (1998). Numerical ecology, 2nd Englished. Elsevier.
Desdevises, Y., Legendre, P., Azouzi, L. and Morand, S. (2003). Quantifying phylogenetic structured environmental variation, Evolution, 57(11):2647-2652
Diniz-filho, J.A.F., Rangel, T.F., Santos, T. and Bini, L.M. (2012). Exploring patterns of interespecific variation in quantitative traits using sequential phylogenetic eigenvector regressions. Evolution, 66(4):1079-1090.
Diniz-filho, J.A.F., Bini, L.M., Rangel, T.F., Morales-Castilla, I., Olalla-Tarraga, M.A., Rodriguez, M.A. and Hawkins, B.A. (2012). On the selection of phylogenetic eigenvectors for ecological analyses. Ecography, 35:239-249.
# NOT RUN {
library(ape)
tree <- rcoal(10)
#Decomposing phylogenetic distance matrix derived from tree into a set of orthogonal vectors
x <- PVRdecomp(tree)
trait <- runif(10)
y <- PVR(x, trait = trait, method = "moran")
str(y)
# }
Run the code above in your browser using DataLab