CODEWITHSUNDEEP
rcluster-analysisheatmap
Gabelins
Gabelins

Reputation: 285

clusters, heatmap and getting info from clusters by R

I've started recently using R to cluster my data. My purpose is to have a heatmap with the related dendrogram and to identify on the heatmap the clusters by squares. So far I tried hclust from gplots package and I could draw rectangles on the dendrogram with the code below:

a <-read.table ("test.txt", header = TRUE)
b <- as.dist(a)
dend <- hclust(b, method = "complete")
plot(dend)
groups <- cutree(dend, k=3)
rect.hclust(dend, k=3, border = "green")

My test.txt file look like this:

      a     b     c     d     e     f
a     1   0.1   0.9   0.5  0.65   0.9
b   0.1     1  0.39  0.83  0.47  0.63
c   0.9  0.39     1  0.42  0.56  0.84
d   0.5  0.83  0.42     1  0.95  0.43
e  0.65  0.47  0.56  0.95     1  0.14
f   0.9  0.63  0.84  0.43  0.14     1

I tried this code that works to obtain a dendrogram and the related clusters. Dedrogram with rectangles from test.txt data.

What I really want is something similar but with the heatmap. I would like the heatmap with squares around the clusters and the lists of the members of the different clusters. The heatmap should look like this:

As I work with large data (a 5200 x 700 matrix), I need a way to save the lists of members from each cluster. I tried also pheatmap from the pheatmap package but I'm not sure about clustering and I cannot have rectangles around clusters.

I'll be very happy to welcome suggestions and comments.

Upvotes: 0

Views: 305

Answers (1)

lawyeR
lawyeR

Reputation: 7674

Are you looking for something like this? I created the rectangle by hand, and you may be looking for R to do so, and for the three clusters above.

data <- read.table(file = "clipboard") # copied your test.txt file from above
data.m <- melt(data)
ggplot(data.m, aes(x=variable, y=value, fill=value)) + geom_tile() + 
  scale_fill_gradient(low="red", high="green") +
  annotate("rect", xmin = min(as.integer(data.m$variable)), xmax = max(as.integer(data.m$variable)),
           ymin = .01, ymax = .2, fill = "transparent", col="black", lwd=2)

heatmap with rectangle

Upvotes: 1

Related Questions