• Intend to run every year going forward
• This one is aimed at beginners
• Spring Into Bioinformatics in Sept-Oct
  • Should have two streams: Beginner & Intermediate
• Organised along with BIG-SA

• Day 1: R & RStudio
  • Very interactive, statistics and results
• Day 2: Bash & The Command Line
  • Large file manipulation, HPC Interaction
• Day 3: NGS Data
  • Focussed on RNA Seq this time
  • Use all skills from Days 1 & 2

• Dan Kortschak, Jimmy Breen, Nathan Watson-Haigh
• Cuong To, Rick Tearle, Lloyd Low, Terry Bertozzi
• Ning Liu, Quanhui Wan, Melanie Smith, Sabrina Ng
• Paul Wang, John Toubia

Steve Pederson

• Co-ordinator, Bioinformatics Hub (University of Adelaide)
• R User for >15 years
• Short training course with Hadley Wickham
• Made countless typos and bad decisions

• Heavily used for analysis of biological data (along with Python)
• Handle extremely large datasets
  
• Huge user base of biological researchers (www.bioconductor.org)
• Easily perform complex analytic procedures
  • Many come as inbuilt functions
    
• Experience is the best teacher

• Avoids common Excel pitfalls
• Makes communicating results simple
• Reproducible Research!!!
  • Transparent methods
  • Code for analysis is equivalent to a lab book
  • Excellent integration with version control such as git

• First appeared in 1993
  • Derivative of S (Chambers et al, Bell Labs 1976)
  • Ross Ihaka and Robert Gentleman
  • Authors wrote for their own research and students
  • Now estimated >2 million users

• Open source
  • R is run by a volunteer community (R Core)
  • John Chambers is still a member of R Core
  • Quarterly release schedule
• RStudio is product of a for profit company
  • An Integrated Development Environment (IDE)
  • Employs many of the best & brightest package developers (tidyverse, bookdown, reticulate, roxygen2 etc)

• Packages are stored on CRAN (https://cran.r-project.org)
  • Packages are a collection of related functions
  • Not curated for statistical quality or documentation
• Bioconductor is a secondary repository (https://www.bioconductor.org)
  • Packages have a biological/genomics focus
  • Curated for consistency & documentation

• Used by Google, ATO, Trump etc
• Very large community of users in finance (Dirk Eddelbuettel - Rcpp)
• Genomics & Ecological Research
• Strong integration with HPC systems like Amazon, Hadoop
• Growing Machine Learning capacity
• Even has it's own peer-reviewed Journal (The R Journal)

With great power comes great responsibility - Uncle Ben

To use R we need to understand a little about:

1. Data Types
   
2. Data Structures

Today we will start with:

1. An Introduction to RStudio
2. Reading data into R
3. Manipulating and cleaning data
4. Visualising data
5. Writing reports using RMarkdown

By the end of the day, we'll be able to

1. Load tabular data (.csv and other 'Excel-type' formats)
2. Clean up datasets
3. Edit data after loading
4. Make amazing looking plots
5. Send reports to collaborators made directly from our code

• R and RStudio are two separate but connected things
  • R is like the engine of your car
  • RStudio is the 'cabin' we use to interact with the engine
  • RStudio even comes with extra flashy features not related to R
• R does all the calculations, manages the data, generates plots
• RStudio helps manage our code, displays the plots plus more

• We can write scripts and execute code interactively
• Auto-completion by hitting the <tab> key
• Use of R Projects to manage code/data for each project/analysis
• We can see everything we need (directories, plots, code, history etc.)
• Integration with markdown, \(\LaTeX\), bash, python, C++, version control software etc.

(There is no need to save your R workspace if asked)

1. File > New Project > New Directory > New Project
   • Enter R_Training as the project name (check capitals & underscores)
   • Create Project
2. New Folder then name it data

• R Projects are simply a wrapper for keeping an analysis organised
  • You can easily navigate to a directory with all scripts and data
• The R Project name will always be the directory name
• Not essential, but are good practice and extremely useful

1. File > New File > R Script
2. Save As Introduction.R

• This is just a text editor.
• We enter our commands here but they are not executed
  • We can keep a record of everything we've done
  • We can also add comments to our code
  • Comments start with the # symbol
• We'll return here later

• Where we execute commands \(\implies\) is essentially the "engine"
• We can execute commands directly in the Console

1 + 1

## [1] 2

2 * 2

## [1] 4

As well as performing simple calculations:

• R has what we call an Environment (i.e. a Workspace)
• We can define objects here, or import data
• Like a workbook in Excel, but much more flexible & powerful
• Don't need all those empty rows & columns like Excel
• R performs calculations & runs processes on these objects

• We need to give it a name, and some data
  • Like naming a spreadsheet in Excel (e.g. Sheet1)
• The process looks like nameOfObject <- data
  • The <- symbol is like an arrow
  • Tells R to put the data in the object

In the Console type:

x <- 5

• We have just defined an object called x
• View the contents of the object x by entering it's name directly in the Console, or by calling print()

x
print(x)

Where have we created the object x?

• Is it on your hard drive somewhere?
• Is it in a file somewhere?

• We have placed x in our R Workspace
• More formally known as your Global Environment

• The Environment is like your desktop
• We keep all our relevant objects here and can save all the objects in your workspace as an .RData object

save.image()

Once an object is in the Environment we can perform calculations on it

1 + x

## [1] 6

x^2

## [1] 25

NB: R is case sensitive

R has a series of inbuilt functions, e.g. sqrt(), log(), max(), min() etc.

• We place an object or value inside the () after the name of a function

sqrt(x)

## [1] 2.236068

log(x)

## [1] 1.609438

• Many inbuilt functions are organised into a package called base
  • Always installed with R
  • Packages group similar/related functions together

?base

• This will take you to the Help pane
  • Click on the underlined word Index at the bottom for a list of functions in the base packages

• Instead of a single value, we can give an R object multiple values
• Known as a vector \(\implies\) like a column in Excel
• R requires all elements of a vector to be the same type

x <- 1:5
x

## [1] 1 2 3 4 5

length(x)

## [1] 5

Functions & calculations work on the entire vector

x + 1
x^2
sqrt(x)
max(x)
sum(x)
mean(x)
sd(x)

Best practice for all analysis is to enter our code in the Script Window

• This is a plain text editor \(\implies\) RStudio will:
  • highlight syntax for us
  • help manage indenting
  • enable auto-completion (it can be slower than your typing)
• Enter code in this window and send it to the R Console
  
• We save this file as a record of what we've done

• We can write comments by starting a line with the #
  • Anything following this symbol will not be executed
  • Can write notes to ourselves and collaborators
  • We can also place this at the end of a line with a comment
• Enter the following in the Script Window (but don't do anything else)

# Create an object called x
x <- 5

# Create an object called x
x <- 5

To send this to the Console:

• Place the cursor on a line then Ctrl+Enter (Cmd+Enter on MacOS), or
• Select the lines using the mouse then Ctrl+Enter (or Cmd+Enter)
• Or after selecting one or more lines you can click the Run button
  • Be careful to select all the correct text though…

As well as creating objects, we can use this to write general code

# I'm not sure. Is x greater than zero?
x > 0

## [1] TRUE

Including comments describing your intention, is highly advisable

RStudio will give us suggestions when we ask it to.

• In either the Console or Script Window type ?bas then hit the <tab> key
  • A whole lot of options will appear
  • Very handy with long variable/function names
  • If you can't quite remember the spelling
  • Sometimes completes when you don't want it to

• Next to the Environment Tab is the History Tab
  • Contains everything executed in the Console
  • Useful for when we've been lazy

Best coding practice is to enter code in the Script Window and execute

In the bottom right are a series of tabs

1. Files: This shows your current working directory
2. Plots: Every time you make a graph it appears here
3. Packages: NEVER CLICK OR UN-CLICK ANYTHING HERE
4. Help: We all use this a lot!

• Every tab can be resized using the buttons on the top right
• Window separators can also be be moved

Help > Cheatsheets > RStudio IDE Cheat Sheet

Page 2 has lots of hints:

• Ctrl + 1 places focus on the Script Window
• Ctrl + 2 places focus on the Console
• Ctrl + 3 places focus on the Help Tab