CODEWITHSUNDEEP
rcluster-analysis
Tal Galili
Tal Galili

Reputation: 25326

is there a way to get a "subtree" from hclust ? (R)

I wish to create a "subtree" from an hclust object.

For example, let's say I have the following object:

a <- list()  # initialize empty object
a$merge <- matrix(c(-1, -2,
                    -3, -4,
                     1,  2,
             -5,-6,
             3,4), nc=2, byrow=TRUE ) 
a$height <- c(1, 1.5, 3,4,4.5)    # define merge heights
a$order <- 1:6              # order of leaves(trivial if hand-entered)
a$labels <- 1:6# LETTERS[1:4]    # labels of leaves
class(a) <- "hclust"        # make it an hclust object
plot(a)                     # look at the result

Now I wish the extract from it the following subtree:

a <- list()  # initialize empty object
a$merge <- matrix(c(-1, -2,
                    -3, -4,
                     1,  2
                ), nc=2, byrow=TRUE ) 
a$height <- c(1, 1.5, 3)    # define merge heights
a$order <- 1:4             # order of leaves(trivial if hand-entered)
a$labels <- 1:4# LETTERS[1:4]    # labels of leaves
class(a) <- "hclust"        # make it an hclust object
plot(a)                     # look at the result

How could I access it?

(I know that cutree could get me the objects of the sub tree, but not create an actual hclust object)

Thanks for any help,

Tal

Upvotes: 4

Views: 2902

Answers (2)

qed
qed

Reputation: 23104

If you have the distance matrix, then you might do something similar to this:

subtree <- function(d, idx) {
  hclust(dist(d[idx, idx]))
}
d <- matrix(rnorm(50 * 50), 50)
s <- subtree(d, sample(1:50, 20))
plot(s)

Upvotes: 0

nico
nico

Reputation: 51640

Not sure this is what you're looking for, but you could

a <- as.dendrogram(a)
branch1 <- a[[1]]
branch2 <- a[[2]]

par(mfrow=c(1,3))
plot(a)
plot(branch1)
plot(branch2)

Upvotes: 6

Related Questions