What is this series about?

• Modern research consists of computing, lots and lots of computing
• This series is designed to introduce researchers to the basics of research computing through R
• Warning: assumes you’ve already installed Rstudio and R
  • If you have not done so please make an appointment with me through the CSC and I would happily assist you :)

A high-level data analysis workflow

Import (data) ➡️ Tidy (data) ➡️ Visualize (data) ➡️ Communicate (findings)

Where we are in the series

1. R: Fundamental Concepts with RStudio

Why R: supported and open source

• free and open source
• built for data analysis and statistics
• huge community and package ecosystem (20,000+ packages on CRAN)

Why R: reproducibility and replicability

Term	Meaning
Reproducibility	Same data, same methods
Replicability	New data, same methods

• A script records every step, which makes your work reproducible

Most scientists report failing to reproduce others’ results, and many their own. Survey of 1,576 researchers (Baker, 2016, Nature).

Motivation 🤔

• Stand on the shoulders of giants: others can only build on your work if they can see what you did
• Efficiency: automate repetitive tasks, get to the important task of thinking
• Scale: work with datasets far too large for a spreadsheet
• Your future self: six months from now, you’ll need to know what you did, too

R vs. RStudio

• R is the engine: the language that does the computing
• RStudio is the dashboard: an interface that makes R easier to drive (an integrated development environment or IDE)

Today’s Goals

By the end of this session, you should be able to:

• Find your way around RStudio
• Install and load packages
• Run code and use functions
• Create and name variables
• Recognize R’s main data types and data structures
• Know where to look for help

RStudio

Installing packages

# install.packages("dplyr") # This call installs a package 
# library(dplyr) # This loads a package
# devtools packages
# install.packages("devtools")
# install_github("username/repositoryname")

Simple Math

2 + 2 # add

## [1] 4

3 - 2 # subtract

## [1] 1

3 * 3 # multiply

## [1] 9

4 / 2 # division

## [1] 2

sqrt(9) # square root

## [1] 3

Functions

# c() is a common function for concatenating things together
c(1:10)

##  [1]  1  2  3  4  5  6  7  8  9 10

Variables

Creating variables

# You can then recall the values (object) associated with your variable...
my_variable <- c(1:10)
myAlphabet <- letters[1:5]
myNumbers <- c(1:5)
#And plug it into functions, ie, do computations on it...

my_variable

##  [1]  1  2  3  4  5  6  7  8  9 10

my_variable * 2

##  [1]  2  4  6  8 10 12 14 16 18 20

Naming variables

# camel case
myNumber <- "Hello" # bad naming
myNumber <- 10 # good naming
# skeleton case
my_number <- 10

Data types

• data (and the variables that we use to store them) come in different shapes and sizes
• think of a variable as a box and the data are the things you want to put in, not all boxes fit all types of stuff (data)
• note: the function typeof(), will tell you what data type you have…

Boolean (Logical Data)

• boolean data are generally referred to as logical

myBool <- TRUE # this is camel case
typeof(myBool)

## [1] "logical"

Complex

myComplexNum <- 9 + 3i # remember sqrt(-1) :)
typeof(myComplexNum)

## [1] "complex"

Vectors

# numeric vector
c(1:10)

##  [1]  1  2  3  4  5  6  7  8  9 10

# character vector
c("a", "b", "c")

## [1] "a" "b" "c"

# logical vector
c(TRUE, TRUE, FALSE)

## [1]  TRUE  TRUE FALSE

# note that vectors do not allow you to mix types
# c(TRUE, 10) # this will convert TRUE to 1
# c(TRUE, "10") # this will convert TRUE to "TRUE"

Matrices

matrix(1:10, nrow = 2)

##      [,1] [,2] [,3] [,4] [,5]
## [1,]    1    3    5    7    9
## [2,]    2    4    6    8   10

Data frames

data.frame(letter = letters[1:5],
           nol = 1:5) # nol mean number of letter

##   letter nol
## 1      a   1
## 2      b   2
## 3      c   3
## 4      d   4
## 5      e   5

Lists

list(item_1 = myAlphabet,
     item_2 = c(1:3))

## $item_1
## [1] "a" "b" "c" "d" "e"
## 
## $item_2
## [1] 1 2 3

Arrays

• A matrix with more than two dimensions: rows, columns, and layers
• Like a stack of ice cube trays or milk carton containers

my_array <- array(1:12, dim = c(2, 3, 2)) # 2 rows, 3 columns, 2 layers
my_array

## , , 1
## 
##      [,1] [,2] [,3]
## [1,]    1    3    5
## [2,]    2    4    6
## 
## , , 2
## 
##      [,1] [,2] [,3]
## [1,]    7    9   11
## [2,]    8   10   12

Identifying Data Structures

class(c(1:10))

## [1] "integer"

class(matrix(1:10, ncol = 2))

## [1] "matrix" "array"

class(data.frame(var_1 = letters,var_2 = 1:26))

## [1] "data.frame"

class(list(item_1 = letters,item_2 = 1:10,item_3 = c(TRUE, FALSE)))

## [1] "list"

Resources: the ? operator (and vignettes)

?sqrt                    # help page for a function
??regression             # search all help pages for a topic
browseVignettes("dplyr") # long-form guides for a package

References

• aRrgh: a newcomer’s (angry) guide to R, by Tim Smith and Kevin Ushey - http://arrgh.tim-smith.us/
• YaRrr! The Pirate’s Guide to R, by Nathaniel D. Phillips - https://bookdown.org/ndphillips/YaRrr/
• R for Graduate Students, by Wendy Huynh. - https://bookdown.org/yih_huynh/Guide-to-R-Book/
• Advanced R, by Hadley Wickham - http://adv-r.had.co.nz/Introduction.html

And through the library, the O’Reilly platform hosts a plethora of titles related to R.

• https://www.w3schools.com/r/r_data_types.asp
• Baker, M. (2016). 1,500 scientists lift the lid on reproducibility. Nature, 533(7604), 452–454. https://doi.org/10.1038/533452a