R course

Daniel Vaulot

2023-01-19

Introduction to R

R sessions

  • 01 - Introduction to R
  • 02 - Data wrangling
  • 03 - Data visualisation
  • 04 - Markdown
  • 05 - Analysis of metabarcoding data

R - Session 01

  • What is R and why use R ?
  • Resources
  • Get started
  • Fundamentals of R
  • Data objects
  • Vectors
  • Operators
  • Functions
  • Packages

Introduction

  • If you are an R guru:
    • Please refrain to answer during this session…
    • Help your neighbor
  • Two special slide formatting
    • Your turn…
    • Warning

Computer languages

History of computer languages

History of R

  • Mid 1970s - S Language for Statistical Computing conceived by John Chambers, Rick Becker, Trevor Hastie, Allan Wilks and others at Bell Labs

  • Early 1990’s - R was first implemented in the early 1990’s by Robert Gentleman and Ross Ihaka, both faculty members at the University of Auckland.

  • 1995 - Open Source Project

  • 1997 - Managed by the R Core Group

  • 2000 - First release of R

  • 2011 - First release of R studio

  • Historical notes - Paper from 1998

Why use R ?

  • Script vs. Menu driven software (e.g. Excel)
    • Can be re-rerun with new data
    • Reproducible workflow
  • Open source
    • Huge number of libraries
    • Tidy “universe” : tidyverse and ggplot2
      • Very easy to manipulate tables (select columns, create new variables)
      • High quality graphics
  • Work environment
    • R studio
  • Document your data processing
    • R markdown
    • Create HTML, pdf, presentations
  • Share your data and workflow
    • GitHub

What can you do with R ?

  • Science
    • Statistics of course…
    • Data processing
    • Graphics
    • Time series analyses
    • Maps
    • Bioinformatics
  • But also
    • Teach
    • Do a presentation
    • Write your CV
    • Build a web site
    • Write a book
    • Much more…

Example of web page

Help

Cheat sheets

Forum

Let’s get started

The R studio interface

  • Bottom left
    • Console
  • Top left
    • File editor for .R and .Rmd files
    • Data frame visualization
  • Top right
    • Environment (i.e. R objects)
    • History
  • Bottom right
    • Files
    • Plots
    • Packages
    • Help

Create a new project

  • Open R studio
  • Create new project for the course in a new directory
    • e.g. Microbes course

Your first script

Two ways to proceed

  1. Type directly in command window
> print("Hello world")
[1] "Hello world"


  1. Create a new script

Type in script window

* Select and execute (CTRL-R)

* Source the script

R objects

Variables

Variables are abstracting your data.

Variables are objects

  • Create a variable
> greeting = "Hello world"
> print(greeting)
[1] "Hello world"
  • Update variable
> greeting = "Bonjour"
> print(greeting)
[1] "Bonjour"

Assignement

  • Assignement done with <-
>  x <- 1
>  y <- 2
>  x + y
[1] 3
>  z <- x + y
>  z
[1] 3
  • = can be used instead of <- but refrain from it (not good style)
>  z = x + y

Visualizing objects

You can view the values of the objects in R-studio environment window (top-right)

R is case sensitive

 z
[1] 3


 Z
Error in eval(expr, envir, enclos): objet 'Z' introuvable

Rules for naming objects

  • Use
    • letters
    • numbers
    • the dot
    • the underscore (not the minus sign !)
  • Start always with a letter

    • Myvariable, Myvariable1, Myvariable.1,Myvariable_01 are OK

    • 1Myvariable, My-variable, Myvariable@ are not OK

Rules for naming objects

  • Use consistent naming: five conventions
    • alllowercase: e.g. adjustcolor
    • period.separated: e.g. plot.new
    • underscore_separated: e.g. numeric_version
    • lowerCamelCase: e.g. addTaskCallback
    • UpperCamelCase: e.g. SignatureMethod
  • Prefer third one, much more easy to read
    • Use names for objects : last_name
    • Use verbs for function : build_name
  • Think about best order
    • e.g. prefer maybe name_last because then you can have name_first, name_full…
    • and you identify that all these objects are related to a name…

Data types

  • character: “Daniel”, “This is a course in R”, ‘Joe Biden’

  • numeric: 2, 15.5, 10e-3

  • integer: 2L (the L tells R to store this is an integer)

  • date: 2018-02-25

  • logical: TRUE, FALSE

  • complex: 1+4i (complex numbers with real and imaginary parts)

  • No data “NA”

  • Not a number “NaN” (e.g. division by zero)

Data structures

  • Vector

  • List

  • Matrix

  • Data frames

  • Function

Vectors

  • The basic R structure is a vector (think as a column in Excel): \[\begin{bmatrix}10 \\ 20 \\ 30 \end{bmatrix}\]
  • A vector can contain only a single element \[\begin{bmatrix}10 \end{bmatrix}\]
  • Assign a value to a vector
 x <- 10
 x
[1] 10

Vectors

  • Assign several elements
 x <- c(10,20,30)
 x
[1] 10 20 30
  • Assign range
 x <- 10:30
 x
 [1] 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
  • Assign characters
 PoTU <- c("Jo", "Biden")
 PoTU
[1] "Jo"    "Biden"
  • Assign logical
 flags <- c(TRUE, FALSE, TRUE)
 flags
[1]  TRUE FALSE  TRUE

Access specific elements of a vector

  • First
 x[1] 
[1] 10
  • Range
 x[1:5] 
[1] 10 11 12 13 14
  • Remove one element
 x[-1] 
 [1] 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

Determine object properties

Apply functions (we will come back to functions latter)

  • typeof() - what is the object’s data type (low-level)?
  • length() - how long is it? What about two dimensional objects?
 typeof(x)
 length(x)
[1] "integer"
[1] 21

What is the type and length of PoTU ?

Operators

Arithmetic Operators

Operator Description
+ addition
- subtraction
* multiplication
/ division
^ or ** exponentiation
x %% y modulus (x mod y) 5%%2 is 1
x %/% y integer division 5%/%2 is 2

We are performing vector operations !

\[\begin{bmatrix} 1\\2\\3\\..\end{bmatrix}+\begin{bmatrix}1\\2\\3\\..\end{bmatrix}=\begin{bmatrix}2\\4\\6\\..\end{bmatrix}\]

Think about it as adding 2 columns in Excel.

Arithmetic Operators

  • Vector one element
 x <- 1
 y <- 2
 z <- x + y
 z
[1] 3
  • Vector several elements
# Two instructions on the same line
 x <- 1:9;  y <- 1:9
 z <- x + y
 z
[1]  2  4  6  8 10 12 14 16 18
  • Several instructions on same line separated by ;
  • The hastag # indicate a comment -> Use heavily to document your code
  • However, it is even better to use R markdown (we will see it later)

Use the other operators

Arithmetic Operators

  • What happens when the vectors have different number of elements ?
 x <- 1:9
 y <- 1
 z <- x + y
 z
[1]  2  3  4  5  6  7  8  9 10

Equivalent to

 y <- c(1,1,1,1,1,1,1,1,1)

The recycling rule…

Can we add logical ?

 x <- TRUE
 y <- FALSE
 z <- x + y
 z
[1] 1

No error but…

The resulting variable is transformed to a numeric

How you would show that ?

 typeof(x)
[1] "logical"
 typeof(z) 
[1] "integer"

Logical Operators

Operator Description
< less than
<= less than or equal to
> greater than
>= greater than or equal to
== exactly equal to
!= not equal to
!x Not x
x | y x OR y
x & y x AND y
isTRUE(x) test if X is TRUE 
 x <- TRUE
 y <- FALSE
 z1 <- x | y
 z2 <- x == y
 z1
[1] TRUE
 z2
[1] FALSE

Do not mix

  • == which is logical operator
  • = which is assignement

Can we add characters ?

 first <- "Jo"
 last <- "Biden"
 full <- first + last
Error in first + last: argument non numérique pour un opérateur binaire

Generates an error

What can we do ?

Functions

Functions

Functions perform specific task on objects

  • e.g. to concatanate strings we use paste()
  paste(first,last)
[1] "Jo Biden"

  • Functions take arguments and return an object called result

  • To know the arguments

    • Use “?”
    • Can also go directly to Help panel and type function name
  ? paste()  # Do not forget the parenthesis

Getting what you want

Let’s apply paste :

  paste(first,last)
[1] "Jo Biden"
  • We would like to get “Jo_Biden”

Can you read the help and suggest a change in the way we call the function ?

  paste(first,last, sep="_")
[1] "Jo_Biden"

Write your own function

If you write 3 times the same piece of code, then write a function…

 my_sum <- function(a, b) {
   c <- a + b
   return (c)
 }
  • my_sum : function name
  • a, b : arguments
  • instructions are enclosed by braces ({})
  • return() : the value(s) returned
  • More compact way
 my_sum <- function(a, b) {a + b}

Call your function

 my_sum(10, 20)
[1] 30
  • better
 my_sum(a = 10, b = 20)
[1] 30

Write a function to compute a product

Examples of functions

Most of the time you do not have to write functions because someone has already written one for what you want to do…

  • Sum
 x <- 1:100
 sum(x)
[1] 5050
  • Sampling a normal distribution
 y <- rnorm(10, mean = 0, sd = 1)
 y
 [1]  0.1262730 -0.4662605 -0.9639646  1.3734205 -1.2810443  1.7808202
 [7]  0.5250865 -0.1196463 -0.8905748  1.0650329

Statistics

 mean(y)
 sd(y)
[1] 0.1149143
[1] 1.051277

Sample more points… 10,000 instead of 100

 y <- rnorm(10000, mean = 0, sd = 1)
 mean(y)
 sd(y)
[1] 0.008637052
[1] 0.9954509

Plot

  • Histogram
  library(graphics)
  hist(y)

What is this “library()”

Packages

Packages

  • Packages are set of functions that have a common goal.

  • They are really the strength of R

  • And these are only the “official”” packages. You can find more on GitHub

Installing a package

Download on your computer the package you need

Install package stringr (to manipulate strings of characters)

Using a package

To use functions from the package

  • use the syntax package::function
 stringr::str_c(first,last, sep= " ")
[1] "Jo Biden"

OR

  • load the package with the library function
 library(stringr)
 str_c(first,last, sep= " ")
[1] "Jo Biden"

Sometimes functions from different libraries have similar names

List installed packages

Recap

  • R is case sensitive: Z != z
  • Objects: data types vs data structures
  • Vectors: think in vector operations
  • Operators: arithmetic vs. logical
  • Functions: try to practice

Next: 02 - Data wrangling

  • Data frames
  • Concept of tidy data
  • Reading data
  • Manipulating data
  • Selecting columns
  • Selecting rows