#load packages
require(vegan)
## Loading required package: vegan
## Loading required package: permute
## This is vegan 2.0-6
require(ade4)
## Loading required package: ade4
## Attaching package: 'ade4'
## The following object(s) are masked from 'package:vegan':
##
## cca
## The following object(s) are masked from 'package:base':
##
## within
require(MASS)
## Loading required package: MASS
#preloaded dataframe
data(varespec)
#Part 1
vare.dis <- vegdist(varespec)
#What was the default dissimilarity index for vegdist()?
vare.iso <- isoMDS(vare.dis)
## initial value 18.026495
## iter 5 value 10.095483
## final value 10.020469
## converged
#create a shepards plot
stressplot(vare.iso, vare.dis)
ordiplot(vare.iso, type = "t")
## Warning: Species scores not available
#Part 2 #We can use raw values for metaMDS function
vare.mds <- metaMDS(varespec)
## Square root transformation
## Wisconsin double standardization
## Run 0 stress 0.1843
## Run 1 stress 0.1948
## Run 2 stress 0.2051
## Run 3 stress 0.195
## Run 4 stress 0.1826
## ... New best solution
## ... procrustes: rmse 0.04174 max resid 0.1523
## Run 5 stress 0.2083
## Run 6 stress 0.1843
## Run 7 stress 0.2092
## Run 8 stress 0.2261
## Run 9 stress 0.1956
## Run 10 stress 0.1967
## Run 11 stress 0.234
## Run 12 stress 0.2178
## Run 13 stress 0.1843
## Run 14 stress 0.187
## Run 15 stress 0.2138
## Run 16 stress 0.212
## Run 17 stress 0.1962
## Run 18 stress 0.2419
## Run 19 stress 0.1948
## Run 20 stress 0.1976
#interpret the output, i.e. why do we transform the data? What is the purpose of the different runs #what is the “procrustes” function doing here? Can we interpret this NMDS with confidence?
ordiplot(vare.mds, type = "t")
#Why do we have species scores this time?
#Choose an add-on, stats or clustering!