Phylogenetic covariance or correlation Matrix at the community level
get_comm_pair_r.RdThe function computes the expected covariances or correations of any community-level properties between communities.
Arguments
- comm
A data frame describing species (column) abundance / presence-absencene in each site, or sample in general (row).
- VCV_sp
A phylogenetic covariance (not correlation!) matrix based on any evolutionary models.
- comm_kronecker
For internal calculation only in
EcoCoMix.- force.PD
Forcing Positive Definite using
nearPD. Defaulted as FALSE.
Details
The function upscales species-level covariance to community-level covariance, which is subsequently converted to a correlation matrix.
Value
- corM
Expected correlation between communities.
- covM
Expected covariance between communities.
- comm_kronecker
For internal uses to speed up the optimization processes.
Examples
library(phytools)
library(tidyverse)
library(EcoCoMix)
library(ape)
library(DHARMa)
data(KSR)
data(KSR_MLtree)
data(KSR_EF)
phy_comm_M <- get_comm_pair_r(KSR,vcv(KSR_MLtree))$corM
## if you prefer other evolutionary models
library(geiger)
KSR_tree_ou <- rescale(KSR_MLtree, model = "OU", alpha = 0.1) #alpha can be other values! see the documentations in geiger for details and other models
plot(KSR_tree_ou)
plot(KSR_MLtree) #compare two trees
phy_comm_M_ou <- get_comm_pair_r(KSR,vcv(KSR_tree_ou))$corM