#####################################
#Load libraries
#####################################
library(vegan)
library(MASS)
library(labdsv)
library(cluster)
library(indicspecies)
#####################################
#Load Data
#####################################
community<-read.csv(file="example.csv", row.names='Sites', sep=",", header=TRUE)
envdata<-read.csv(file="example_env.csv", row.names='Sites', sep=",", header=TRUE)

# Perform indicator species analysis based on spatial grouping (spatial variable with levels a,b or c)
ind_species<-multipatt(community,envdata$spatial,max.order=3,duleg=TRUE,func="IndVal.g",control=how(nperm=5000))
ind_species
summary(ind_species)

# Same as above but set max.order=2 and duleg=F to consider combinations of groups
ind_species<-multipatt(community,envdata$spatial,max.order=2,duleg=FALSE,func="IndVal.g",control=how(nperm=5000))
ind_species
summary(ind_species)

# indicator power of species
# classify sites based on the presence of species 3
classification<-community$sp3>0
ind_species<-multipatt(community,classification,duleg=TRUE)
ind_species
summary(ind_species)

# use indpower function in vegan package
indpower(community)

# SIMPER
sim<-simper(community,envdata$spatial)
sim
summary(sim)