R Programming for Biologists: Workshop 3

Success?

How could you calculate the mean of each element in the below list?

my_list <- list("a"=c(15,14,526),
                "b"=c(417,6541,14), "c"=c(5,4))

Using lapply is easier!

lapply(my_list, mean)

$a
[1] 185

$b
[1] 2324

$c
[1] 4.5

my_df <- data.frame(matrix(c(1,5,7,4,1,5,7,3,14), 
                  nrow=3, ncol=3))

lapply(my_df, mean)

$X1
[1] 4.333333

$X2
[1] 3.333333

$X3
[1] 8

x <- 1:4
lapply(x, runif)

[[1]]
[1] 0.04721432

[[2]]
[1] 0.5410313 0.4316071

[[3]]
[1] 0.2415562 0.6597251 0.4966759

[[4]]
[1] 0.9568958 0.3634965 0.8942512 0.6853268

x <- 1:4
lapply(x, runif, min=0, max=10)

[[1]]
[1] 2.047823

[[2]]
[1] 1.5909216 0.6347786

[[3]]
[1] 7.711469 6.559685 3.702482

[[4]]
[1] 7.6326175 0.8124635 0.6938888 9.7873019

A function that is unnamed and only defined in the loop.

my_list <- list("a"=c(15,14,526),
                "b"=c(417,6541,14), "c"=c(5,4))

lapply(my_list, function(x) { x[1] + x[2] })

$a
[1] 29

$b
[1] 6958

$c
[1] 9

my_df <- data.frame(matrix(c(1,5,7,4,1,5,7,3,14), 
                  nrow=3, ncol=3))
my_df

  X1 X2 X3
1  1  4  7
2  5  1  3
3  7  5 14

apply(my_df, 2, mean)

      X1       X2       X3 
4.333333 3.333333 8.000000 

apply(my_df, 1, sum)

[1] 12  9 26

If summing or averaging rows or columns, you should use these specialized functions:

• rowSums
• colSums
• rowMeans
• colMeans

x <- matrix(rnorm(200), 4, 50)
apply(x, 1, quantile, probs=c(0.25, 0.75))

          [,1]       [,2]       [,3]       [,4]
25% -0.8391107 -0.5240767 -0.7099116 -0.8795485
75%  0.4337546  0.8268244  0.8929874  0.4903421

• sapply: returns output of lapply as vector if possible.
• mapply: applies function in parallel to set of arguments.
• tapply: applies function over subsets of a vector.

• base –> basic plots
  
• lattice –> focus on breaking plots down by facets
  
• ggplot2 –> popular and powerful (but steeper learning curve)

• plot

• boxplot

• hist

• etc…

data(iris)
plot(iris$Sepal.Length, iris$Petal.Length, col=iris$Species, pch=16)

boxplot(iris$Sepal.Length ~ iris$Species, outline=FALSE, col="grey")

library(beeswarm)
boxplot(iris$Sepal.Length ~ iris$Species, outline=FALSE, col="grey")
beeswarm(iris$Sepal.Length ~ iris$Species, pch=16, add=TRUE)

hist(iris$Petal.Length, col="grey", breaks=50)
abline(v=2.5, lwd=2, lty=2, col="black")

Useful when you have multiple factors and you want to compare the levels (or facets).

data(CO2)
summary(CO2)

     Plant             Type         Treatment       conc     
 Qn1    : 7   Quebec     :42   nonchilled:42   Min.   :  95  
 Qn2    : 7   Mississippi:42   chilled   :42   1st Qu.: 175  
 Qn3    : 7                                    Median : 350  
 Qc1    : 7                                    Mean   : 435  
 Qc3    : 7                                    3rd Qu.: 675  
 Qc2    : 7                                    Max.   :1000  
 (Other):42                                                  
     uptake     
 Min.   : 7.70  
 1st Qu.:17.90  
 Median :28.30  
 Mean   :27.21  
 3rd Qu.:37.12  
 Max.   :45.50  

library(lattice)
histogram(~ conc | Type + Treatment, data=CO2)

xyplot(uptake ~ conc | Type + Treatment, data=CO2)

library(ggplot2)
data(iris)
ggplot(data=iris, aes(x = Sepal.Length, y = Sepal.Width)) + geom_point()

library(ggplot2)
data(iris)
my_plot <- ggplot(data=iris, aes(x = Sepal.Length, y = Sepal.Width)) + geom_point()

my_plot <- my_plot + theme_bw() + geom_point(aes(color=Species, shape=Species)) + xlab("Sepal Length") +  ylab("Sepal Width")

Many plot types both in base and ggplot2 are shown here on the iris dataset: https://www.mailman.columbia.edu/sites/default/files/media/fdawg_ggplot2.html

The 3 “Simulation” videos. I'll cover these concepts along with basic stastitical tests and how to do build basic models in R.