This document is a work in progress, please give us feedback

1 Functions

Functions allows you to perform a specific task in a more powerful and general way. When you need to run the same thing repeatedly, the most efficient and reproducible way is to write a function. Functions will avoid incidental mistakes when you copy and paste code in different parts of your script.

Let’s look at the following example of code which rescales each column of a data frame to a range from 0 to 1

 df<-data.frame(
   a=rnorm(10),
   b=rnorm(10),
   c=rnorm(10)
 )

df$Rescale_a<-(df$a-min(df$a,na.rm=TRUE))/(max(df$a,na.rm=TRUE)-min(df$a,na.rm=TRUE))
df$Rescale_b<-(df$b-min(df$b,na.rm=TRUE))/(max(df$b,na.rm=TRUE)-min(df$b,na.rm=TRUE))
df$Rescale_c<-(df$c-min(df$c,na.rm=TRUE))/(max(df$c,na.rm=TRUE)-min(df$c,na.rm=TRUE))

df$Rescale_a
##  [1] 0.5266659 0.1690370 0.3059742 0.3205623 0.2578741 0.2164492 1.0000000
##  [8] 0.2427424 0.5413472 0.0000000

1.1 Basic structure

function_name <- function(inputs) {
  output_value <- do_something(inputs)
  return(output_value)
}

Let’s try our first function called rescale01

rescale01<-function(x){
  
  rescale_value<-(x-min(x, na.rm=TRUE))/(max(x,na.rm=TRUE)-min(x,na.rm=TRUE))
  return(rescale_value)
  
}

rescale01(x=df$a)
##  [1] 0.5266659 0.1690370 0.3059742 0.3205623 0.2578741 0.2164492 1.0000000
##  [8] 0.2427424 0.5413472 0.0000000
rescale01(df$a)
##  [1] 0.5266659 0.1690370 0.3059742 0.3205623 0.2578741 0.2164492 1.0000000
##  [8] 0.2427424 0.5413472 0.0000000
df$Rescale_a<-rescale01(df$a)

We can combine functions to do larger tasks in two ways. In this case, we can use the function range() which gives us the minimum and maximum values directly

rescale01<-function(x){
  
  ran<-range(x, na.rm=TRUE)
  rescale_value<-(x-ran[1])/(ran[2]-ran[1])
  
  return(rescale_value)
  
}

rescale01(df$a)
##  [1] 0.5266659 0.1690370 0.3059742 0.3205623 0.2578741 0.2164492 1.0000000
##  [8] 0.2427424 0.5413472 0.0000000
df$Rescale_a<-rescale01(df$a)

1.2 Walk through function execution

  • Call function rescale01
  • Assign the vector df$a to the x argument (or input)
  • Calculate the range of the vector
  • Using the min and max values rescale the vector and store the new values in the rescale_value object
  • Send the rescale_value object back as output
  • Store it in df$Rescale_a

You can also write the function in a simpler way.

rescale01<-function(x){
  
  (x-min(x, na.rm=TRUE))/(max(x,na.rm=TRUE)-min(x,na.rm=TRUE))

}

1.3 Challenge

  1. Write a function that calculates the volume of a plant using lenght, width, height as input arguments. volume=length * width * height
  2. Calculate the volume for the following plants
Plants length (in) width (in) height (in)
A 20.32 40.64 50.80
B 25.40 38.10 60.96
C 25.4 50.8 127

  1. Calculate a function that converts inches in centimeters (there are 2.54 cm in one in). Use that function to transform the lenght, width, height plant values.

  2. Create a new function that calculates the volume of a plant in cm^3 using a combination between the functions in 1 and 3.

2 Conditionals

Conditionals evaluate whether a logical statement is either TRUE or FALSE. Conditional statements use operators such as ==, !=, <,>, <=, >=. You can combine these with & (and) or | (or). You can also use some predefine functions such as identical()

20>7
## [1] TRUE
"A"=="B"
## [1] FALSE
"A"!="B"
## [1] TRUE
identical("C","D")
## [1] FALSE

2.1 Challenge

Taken from data carpentry[http://www.datacarpentry.org/semester-biology/exercises/Making-choices-choice-operators-R/]

  1. Create the following variable
w <- 10.2
x <- 1.3
y <- 2.8
z <- 17.5
dna1 <- "attattaggaccaca"
dna2 <- "attattaggaacaca"
  1. Use them to print whether or not the following statements are TRUE or FALSE
  • w is greater than 10
  • w + x is less than 15
  • x is greater than y
  • 2 * x + 0.2 is equal to y
  • dna1 is the same as dna2
  • dna1 is not the same as dna2
  • w is greater than x, and y is greater than z
  • x times w is between 13.2 and 13.5
  • dna1 is longer than 5 bases (use nchar() to figure out how long a string is), or z is less than w * x

2.2 Floating point

x<-sqrt(2)^2
x
## [1] 2
x==2
## [1] FALSE

Floating points are unexpected results from computer arithmetics in numbers that contain very small trailing digits.

# Shows the number in a C-style format
formatC( x, format='f', digits=20)
## [1] "2.00000000000000044409"

To avoid floating-point issues, use the function dplyr::near() or round the values of the variable

require(dplyr)

near(x, 2)
## [1] TRUE
round(x)==2
## [1] TRUE

2.3 if statements

if statements allows you to conditionally execute code

if(condition){
  # code executed when condition is TRUE
}else{
  # code executed when condition is TRUE
}

2.3.1 One condition example

Size_tree<-function(x){
  
  if(x>5){
    print("Large tree")
  }else{
    print("Small tree")
  }

}

Size_tree(100)
## [1] "Large tree"
Size_tree(3)
## [1] "Small tree"

2.3.2 Multiple conditions

Size_tree<-function(x){
  
  if(x>5){
    print("Large tree")
  }else if(x<=5 & x>3 ){
    print("Medium tree")
  }else{
    print("Small tree")
  }

}

Size_tree(100)
## [1] "Large tree"
Size_tree(5)
## [1] "Medium tree"
Size_tree(2)
## [1] "Small tree"
in_to_cm<-function(x){
  x*2.54
}

PlantVolume_in_m<-function(l, w, h, units="cm"){
  
  if(units=="in") {
    
    new_l<-in_to_cm(l)/100
    new_h<-in_to_cm(h)/100
    new_w<-in_to_cm(w)/100
    volume <- new_l*new_h*new_w
    
  }else if (units=="cm"){
    
    new_l<-l/100
    new_h<-h/100
    new_w<-w/100
    volume <- new_l*new_h*new_w
    
  }else if (units=="m"){
    
    volume <- l*h*w
    
  }else{
    
    message("Use either m, cm or in")
    volume<-NULL
  }
  
  return(volume)
  
}

PlantVolume_in_m(0.254,0.2794,0.3048,units="m")
## [1] 0.02163092
PlantVolume_in_m(10,11,12,units="in")
## [1] 0.02163092
PlantVolume_in_m(25.4,27.94,30.48,units="cm")
## [1] 0.02163092
PlantVolume_in_m(25.4,27.94,30.48,units="km")
## Use either m, cm or in
## NULL

2.4 Challenge

Go to this link and complete the assignment from Data carpentry for Biologist website

3 Sources

  • Hadley Wickham and Garrett Grolemund. 2017. R for Data Science: Import, Tidy, Transform, Visualize, and Model Data (1st ed.). O’Reilly Media, Inc.

  • Data carpentry for Biologist. Functions in R