Chapter 10 Lab 3 NCI60 Data Example.R

# Chapter 10 Lab 3: NCI60 Data Example

# The NCI60 data

library(ISLR)
str(NCI60)
nci.labs <- NCI60$labs
nci.data <- NCI60$data
dim(nci.data)
nci.labs[1:4]
table(nci.labs)

# PCA on the NCI60 Data

pr.out <- prcomp(nci.data, scale=TRUE)
Cols <- function(vec){
  cols <- rainbow(length(unique(vec)))
  return(cols[as.numeric(as.factor(vec))])
}

par(mfrow=c(1,2))
plot(pr.out$x[,1:2], col=Cols(nci.labs), pch=19,xlab="Z1",ylab="Z2")
plot(pr.out$x[,c(1,3)], col=Cols(nci.labs), pch=19,xlab="Z1",ylab="Z3")
summary(pr.out)
plot(pr.out)
pve <- 100*pr.out$sdev^2/sum(pr.out$sdev^2)

par(mfrow=c(1,2))
plot(pve,  type="o", ylab="PVE", xlab="Principal Component", col="blue")
plot(cumsum(pve), type="o", ylab="Cumulative PVE", xlab="Principal Component", col="brown3")

# Clustering the Observations of the NCI60 Data

sd.data <- scale(nci.data)
par(mfrow=c(1,3))
data.dist <- dist(sd.data)
plot(hclust(data.dist), labels=nci.labs, main="Complete Linkage", xlab="", sub="",ylab="")
plot(hclust(data.dist, method="average"), labels=nci.labs, main="Average Linkage", xlab="", sub="",ylab="")
plot(hclust(data.dist, method="single"), labels=nci.labs,  main="Single Linkage", xlab="", sub="",ylab="")

hc.out <- hclust(dist(sd.data))
hc.clusters <- cutree(hc.out,4)
table(hc.clusters,nci.labs)
par(mfrow=c(1,1))
plot(hc.out, labels=nci.labs)
abline(h=139, col="red")
hc.out
set.seed(2)
km.out <- kmeans(sd.data, 4, nstart=20)
km.clusters <- km.out$cluster
table(km.clusters,hc.clusters)
hc.out <- hclust(dist(pr.out$x[,1:5]))
plot(hc.out, labels=nci.labs, main="Hier. Clust. on First Five Score Vectors")
table(cutree(hc.out,4), nci.labs)