Sorting vertex and adding colors to groups in igraph r

Question

I have an probably really simple problem. I have a matrix of distances MD (see example below) between a group of 6 people. They are sorted into three groups:

ClassA <- c("TOM","EVE","STEVE")
ClassB <- c("JANE","HANNAH")
ClassC <- c("ADAM")

matrix:

           TOM        JANE     ADAM      EVE      STEVE    HANNAH  
 TOM     0.0000000 0.8383965 1.1979821 1.186857 1.3500972 1.125554 
 JANE    0.8383965 0.0000000 1.2319391 1.274157 1.3964041 1.623064 
 ADAM    1.1979821 1.2319391 0.0000000 1.278604 0.8377628 1.432721 
 EVE     1.1868572 1.2741569 1.2786041 0.000000 1.4310217 1.328200 
 STEVE   1.3500972 1.3964041 0.8377628 1.431022 0.0000000 1.465991 
 HANNAH  1.1255540 1.6230643 1.4327214 1.328200 1.4659912 0.000000 

Now I want to plot a minimal spanning tree:

g <- graph.adjacency(as.matrix(MD), weighted=TRUE)
V(g)$label <- V(g)$name 
mst <- as.undirected(minimum.spanning.tree(g)) 

My question is: how to add speciefied colors to verticles belonging to the same group? It's quite simple in the example,

V(g)[1]$color <- "red"
V(g)[2]$color <- "blue"
V(g)[3]$color <- "green"

however, the complete matrix has over 100 pupils, therefore I need something more efficient. I tried to sort it, but my function crashed.

Thank you MrFlick! It's really simple yet effective. Do you perhaps know how to add legend in rglplot? I use it because mst graphs looks better in rgl than in plot, however its main purpose are 3D plots and unfortunately I haven't found any informations how to add legend to 2D plots.

> library(rgl)

> rgl.open()

> rgl.bg(sphere=FALSE, color=c("white"), lit=FALSE)

> rglplot(mst,vertex.size = 3, layout = myLayout, vertex.label.font=10, vertex.label.dist=0.2) 

Answer 1

First, it would be nicer if you had your name/class assignments in a data.frame. (It also would have been nice if they were stored in a list or something). Here I gather all the ClassA/B/C/* varaibles into a list and stack them

indclass <- stack(mget(ls(pattern="Class*")))
indclass
#   values    ind
# 1    TOM ClassA
# 2    EVE ClassA
# 3  STEVE ClassA
# 4   JANE ClassB
# 5 HANNAH ClassB
# 6   ADAM ClassC

and i'll choose colors for each of the class names (the values from the "ind" column above)

cols <- c(ClassA="green", ClassB="lightblue", ClassC="purple")

Then, i'll use match() to look up the class for each vertex and assign a color based on class name

V(g)$color <- cols[ indclass$ind[match( V(mst)$name, indclass$values ) ] ]

Then if you plot your minimal spanning tree, you get

mst <- as.undirected(minimum.spanning.tree(g)) 
plot(mst)