[R] on NMDS graphics

[Gavin Simpson]

On Wed, 2010-12-08 at 13:54 +0530, Sinu P wrote:
> Hi,
> 
> I have used Vegan to construct an NMDS ordination plot. I plotted sites of
> three forest types with the site number in it. My reviewer has asked me to
> use different symbols for each of the forest types.
> Can anyone send me how I can do this in R in simple steps. I have used the
> options like ordiplot, sel and pl syntaxes that are not working for the
> question that I asked for.
> 
> Best,
> Sinu

The commonly used procedure is to set up a vector of colours or symbols
you want to use, one per group. Then set up a variable that contains the
group identifier, as a factor. We then index the colours or symbols
vector with our grouping factor to produce the correct length and
specification of parameters col or pch for the plot.

Here is an example:

require(vegan)
data(varespec)

## Ordination
ord <- metaMDS(varespec)

## Dummy forest types, indicator
## stored as a factor
ftypes <- gl(3, nrow(varespec)/3)
## or via something like
ftypes <- factor(rep(paste("Type", 1:3), each = 8))

## plot the data
plot(ord, display = "sites", type = "n")

## add the points coloured by ftype
## 1) specify colours wanted for groups
vcols <- c("red","green","blue")
## or 1) specify symbols for groups
vpch <- c(16,17,18)
## 2) add points using ftypes to index vcols
points(ord, col = vcols[ftypes], pch = vpch[ftypes])

To see what is being done, try:

> vcols[ftypes]
 [1] "red"   "red"   "red"   "red"   "red"   "red"   "red"   "red"   "green"
[10] "green" "green" "green" "green" "green" "green" "green" "blue"  "blue" 
[19] "blue"  "blue"  "blue"  "blue"  "blue"  "blue"

HTH

G
-- 
%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%
 Dr. Gavin Simpson             [t] +44 (0)20 7679 0522
 ECRC, UCL Geography,          [f] +44 (0)20 7679 0565
 Pearson Building,             [e] gavin.simpsonATNOSPAMucl.ac.uk
 Gower Street, London          [w] http://www.ucl.ac.uk/~ucfagls/
 UK. WC1E 6BT.                 [w] http://www.freshwaters.org.uk
%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%