===================================================
Introductory Biostatistics & Biostatistics using R
** BI217: Problem Set 1
===================================================
1. Given a vector x=c(1,7,2,4,5,3), print the following results:
(1) rank(x); (2) order(x); (3) sort(x); (4) unique(x)

2. if x=c(-4.7, 3.5, 0.7, 3.3, -1.4), then
(1) round(x); (2) ceiling(x); (3) floor(x); (4) quantile(x)

3. Tell what the following R codes can do.
(1) normalization
m <- 10; n <- 5;
x <- matrix(rnorm(m*n), m, n)
colmeans <- apply(x, 2, mean)
colsds <- apply(x, 2, sd)
x <- sweep(x, 2, colmeans, "-")
x <- sweep(x, 2, colsds, "/")

(2) merging
days <- c("Sun", "Mon", "Tue", "Wed", "Thu", "Fri", "Sat");
days1 <- sample(days)
days2 <- sample(days)
x1 <- rnorm(7)
names(x1) <- days1
x2 <- runif(7)
names(x2) <- days2
m1 <- match(days, days1)
m2 <- match(days, days2)
x <- cbind(x1[m1], x2[m2])
rownames(x) <- days
x

(3)
x <- rnorm(1000)
y <- rnorm(1000)
cols <- ifelse(y < -x^2 | y > x^2, 'lightblue', "darkgray")
pchs <- ifelse(y < -x^2 | y > x^2, '+', "x")
plot(x,y,type="n")
points(x,y,col=cols, pch=pchs)
curve(-x^2, lty=2, add=T, lwd=2)
curve(x^2, lty=2, add=T, lwd=2)

4. On the standard normal curve, label the 95%CI of the mean:
(1) two-tail; Hint: qnorm(0.025); qnorm(0.975)
(2) lower-tail; Hint: qnorm(0.95)
(3) upper-tail; Hint: qnorm(0.05)

5. There are some values you should remember, which are
(1) qnorm(0.975), qnorm(0.025)
(2) qt(0.95)

6. The file "STAT3.bed" contains coordinates for the ChIP-seq data mapping to
the mouse genome, which correspond to the possible targets for the transcriptional 
factor --- STAT3.
(1) Import the data into R using the command read.table();
(2) Count the number of incidences for each genome with the table() command;
(3) Which of the genomes have the maximum incidences?