Welcome to R Café

About the course

Welcome to “R-cafe: R for reproducible research”

In this course, the following topics will be covered

Getting started with R and Git - Introduction to working with RStudio, R and Github
R programming with tidyverse - Introduce important tools for data science in R
Markdown/Quarto and `gtsummary` - introduce gtsummary for analysis and creating report with Markdown/Quarto

About the course

By the end of this course, you will be able to conduct an end-to-end data analysis process on a practical dummy dataset, and create a report using R and RMarkdown/Quarto

Teaching team:

Hai Ho Bich (haihb@oucru.org)
Thinh Ong Phuc (thinhop@oucru.org)
Tuyen Huynh Ngoc (tuyenhn@oucru.org)
Anh Phan Truong Quynh (anhptq@oucru.org)
Hung Tran Thai (hungtt@oucru.org)

Set up

Prerequisites

Before this lesson, you should have:

R and RStudio installed
a Github account, Github Desktop installed

Overview

Note

Brief introduction to Git, GitHub, and GitHub Desktop
Intro to RStudio
R project (create R project, set up renv)
Intro to R

Git and GitHub

Git vs. GitHub

Git is a version control system, its purpose is to keep track of changes of files within a repository and provide the capability to revert to previous version
GitHub is a cloud-based service to host Git repository, its main purpose is to store and/or share your Git project with collaborators

Local vs. Remote

Local means your physical computer. When we say local we are talking about what’s happening directly on your computer
Remote can mean a computing cloud, a server, or someone else’s computer. When we say remote we are talking about what’s happening not on your computer, i.e. remotely.

Git and GitHub

Git is a software that works locally. You don’t need GitHub for Git to work. You can use Git as-is
GitHub acts as a remote server for Git, so that you can share your project and collaborate with others remotely
A repository is a place where Git tracks all of your changes to files and folders. GitHub stores a copy of your repository whenever you push to it

Git and GitHub

Terminologies

Branch: a separate environment, identical copy of where and when you branch from
Commit: packaging all the changes you made into a “version”
Push: pushing a commit (a “version”) to remote, e.g. GitHub
Pull: updating your local repository with changes from remote
Add: chosing which changes you want to commit

Git and GitHub

Takeaway points

Git is a standalone software, which works fully locally. You can use Git without GitHub
GitHub is a platform for collaboration. It uses Git as the version control system

GitHub Desktop

Git is a command-line interface (CLI) software.

Caution

We strongly recommend to not use Git through its CLI. It is not fun and you will lose your work because of a typo.

GitHub Desktop

GitHub Desktop is a software that makes it easier for you to use Git. It is a software built by GitHub, again, separate from Git
Advantages:
- Simplifies everything and only provide commonly used functionalities of Git
- Reduces the chances that you will make errors in using Git
- Direct integration to GitHub, makes it easier for authentication

Setting up GitHub Desktop for R Café

Setting up GitHub Desktop

Checklist

Sign in to your account in GitHub Desktop
Clone R Cafe into your computer/laptop. Link
Create your own branch
Publish (push) your branch

Refer to the handouts for instructions

R and RStudio

R is a programming language for statistical computing and data visualization.

RStudio is a software designed to make working with R easier by helping you create, edit, and manage R code and projects. More formally, it is known as an Integrated Development Environment (IDE).

R and RStudio

Why use R?

Cost: R is free and cross-platform.
Reproducibility: Programming in R (as opposed to point-click tools such as Excel/Access) enhances reproducibility, error detection, and reduces manual workload.
Community: R has a large, collaborative user base, with new tools and packages for analysis developed constantly.

R project

Create Project in RStudio

An R project is a directory with .Rproj file, signaling RStudio to manage the project settings accordingly.

The process of creating an R project is as followed

In the menu File > New Project…
Click Existing Directory > Browse
Click Browse and click on your project directory
Click Create Project

RStudio interface

Consists of 4 main panes

Source code: Show the content of a source code file
Console: Show the executed code lines and their output
Defined variables: Show the currently defined variables
Files/Plot/Package/Help:
- Files: for files nagivation
- Plots: show plot output
- Packages: show all the installed packages and packages being used (packages in use will have a ✓)
- Help: show documentations for functions or packages

Project structure

A minimal R project structure will have the following format

└── my_project
    ├── output
    ├── data
    │   ├── raw
    │   └── processed
    └── analysis.R

Where

data folder contains data to be analyzed
output stores code output (plot, figures, etc.)
analysis.R is the file containing R code. There can be multiple .R files under 1 project.

Naming convention

Note

Some rules of thumb for naming files and folders:

Do not use diacritics (e.g. Vietnamese accent marks).
Do not use overly long names, keep it short but meaningful.
Do not include special characters such as /, \, &, ,, (, )... in names. It’s best to use only letters and numbers.
Avoid mixing uppercase and lowercase letters arbitrarily. If necessary, use naming conventions like camelCase (capitalize the first letter of each word after the first one, e.g., dataSoiHcm) or PascalCase (capitalize the first letter of all words, e.g., DataSoiHcm).
Do not include spaces (" ") in names. Instead, you can use camelCase, or replace spaces with - or _. For example, instead of naming a file data soi HCM 2023.xlsx, you could name it dataSoiHcm2023.xlsx, data-soi-hcm-2023.xlsx, or data_soi_hcm_2023.xlsx.

Project environment

Set up renv

To use renv for your project. Run the following line of code in your console

renv::init()

Which will create renv.lock and renv folder under your project

Why use renv?

Encapsulate the development environment of your R projects (R version, packages used and their versions) in a file called renv.lock
Ensure collaborators to replicate your R project’s development environment by simply running renv::restore()
Isolate installed packages under project directory for more convenient management. With renv set up, the installed packages are stored under
- path_to_project/renv/library/{R-version}/{device-architecture}/

What are packages?

An R package is a shareable bundle of functions. We will frequently use available packages to assist us during coding process.

R foundations

R script

To create an R script in RStudio, in the menu section, click File > New File > R Script

R script is where you store your R code and comments.

Comments are notes to explain your code to other developers (or your future self, when you revisit the code). To comment in R simply add # at the start of your note.

R script

Example: Some R codes and comment

# this is a comment, and the following code will print Hello world in the console
print("Hello world")

[1] "Hello world"

# you can also perform computations in R
# In R, you can denote 
# - multiplication as *
# - power as **
# - devision as /
# - e5 is the shorthand notation for *10**5
12*10e5 + 313**2 - 11890/2

[1] 12092024

Variable

In R, a variable is a name used to store an object, such as a dataset, vector, number, or graph. Once assigned to a variable, the object can be easily referenced and reused in future commands.

In R, a variable can be created by assigning an object to it using <-

As previously mentioned, an object can be a number, string, vector of characters, etc.

# assigning 3.14 to the variable pi_numb
pi_numb <- 3.14
# which can then be referenced 
pi_numb

[1] 3.14

# object can also be a text/characters
greeting <- "Hello World" # note that it must be contained within quotation marks
greeting

[1] "Hello World"

Variable

Value of a variable can also be overridden

var <- 5
var

[1] 5

# override variable
var <- "new value"
var

[1] "new value"

Naming variable

The rule for naming variable is similar to that of naming file/folder names (See Section 6.5)

Additionally, variable name needs to be concise and beware of overriding values

Classes

All the objects stored in R have a class which tells R how to handle the object (to check the class of a variable, run class(variable)).

There are many possible classes, but common ones include

Classes	Explanation	Examples
`numeric`	for numerical data, including decimals	`3.14`, `-1000`, `10`, `2e6`, etc.
`logical`	for boolean data (`TRUE`/`FALSE`) without quotation marks	`TRUE` or `FALSE`
`character`	for text/characters data (also referred to as string)	`"Hello world"`, `"R Cafe"`, etc.
`Date`	data classified as Date can be displayed and manipulated in special ways	`2024-12-09` , `s`, etc.
`factor`	for categorical data	gender: female or male, economic status, district, ethnicity, etc.
`data.frame`	common way for R to store dataset. Consists of vectors (columns, each column must be of a specific class) of data bounded together.
`tibble`	an “enhanced” version of `data.frame` which is less prone to error and printed more nicely in the console

Special values

Value	Description
`NA`	Stands for Not Available. Indicates missing values.
`NULL`	Indicates undefined/abscence of value. (e.g. `numb <- NULL` indicates value for variable `numb` is undefined)
`NaN`	Stands for Not a Number. Indicates unrepresentable number (e.g. result for `0/0` )
`Inf`/`-Inf`	Infinity/Negative infinity

Functions

A function is a reusable sequence of code.

It can be created using the syntax

function_name <- function(){

# function content

}

Functions

Example: a simple function

print_hello <- function(){
  print("Hello world")
  print("This is R Cafe course")
}

# defined function can then be called as followed
print_hello()

[1] "Hello world"
[1] "This is R Cafe course"

Functions

Functions can also have arguments, which creates reusable code with dynamic inputs and outputs

A function with arguments as followed. Note that you can include as many arguments as you want

function_name <- function(argument_1, argument_2){
  # function content
  # values of argument can be accessed simply calling argument name
}

Functions

Example: simple function

compute_power <- function(base, power){
  # return `base` raised to the power of `power`
  base**power
}

# output can be assigned to a variable
output <- compute_power(base = 5, power = 2)
output

[1] 25

Functions

Implicit vs. explicit return

By default, R will implicitly return the result of the last line of code in the function
You can also explicitly specify the returned value using return()

compute_power <- function(base, power){

  # explicit return statement
  return(base**power)
}

# output can be assigned to a variable
output <- compute_power(base = 5, power = 2)
output

[1] 25

Functions

Note

You can provide value for each argument without specifying the name but it must follow the same order in the function definition.

However, it is generally recommended to include the name of the argument for better readability. When argument name is specified, order no longer matter.

compute_power(5, 2)

[1] 25

compute_power(2, 5) # expect different result (2^5)

[1] 32

compute_power(base = 5, power = 2)

[1] 25

compute_power(power = 2, base = 5) # expect same result

[1] 25

Functions

When to create a function?

A general rule is when you start to copy-paste a sequence of code multiple times (usually >3), you should make a function for it

The in-class excercise (Section 9) will demonstrate one use case for function

Functions

Some criteria for a good function

Modularity - each function should focus on one task (e.g. you can have a function to import data, another to model the data, another to visualize data but not a function doing all of those 3 tasks)
Concise - the content of a function should not be too lengthy. If a function starts to have > 30 lines of code, it may be an indication that the modularity rule is violated.
Readability - function’s name should clearly describe its purpose (e.g. compute_min, plot_case, etc.), and the content should be properly commented and documented.

Functions

R is a functional programming language, which means you will encounter a lot of functions while coding in R.

To assist users, R also include a built-in documentation system that explains the purpose and usage of each function. To look up a function, use ? or help()

# look up documentation for `sqrt` function
?sqrt

starting httpd help server ... done

help(sqrt)

Note that the built-in documentation only helps when you already know which function you want to ask about. If you are looking for a function to achieve a specific task, it is better to start with search engine (Google, Bing, etc.), forums/Q&A sites (Github discussions, StackOverflow, etc.) or even ChatGPT.

Functions

The documentation that show up when calling help() or ? are usually generated from something called roxygen documentation.

To create a roxygen documentation, place the cursor within the function and from the menu, click Code > Insert Roxygen Skeleton

#' Compute power
#' 
#' Takes two arguments, `base` and `power` then return base raised to given power
#'
#' @param base - a numeric value for base number
#' @param power - a numeric value for power
#'
#' @return numeric value for base^power
#' @examples
#' compute_power(3,4)
compute_power <- function(base, power){
  # return `base` raised to the power of `power`
  base**power
}

compute_power(5, 2)

[1] 25

Function `set.seed()`

set.seed() is essential to ensure reproducibility for processes that uses a random number generator

A seed is an input value to generate the “random data”. By specifying the same seed, we can ensure the number generator always return the same result.

# Set seed 
set.seed(123)

# Generate sample data
sample_data <- rnorm(10, mean = 50, sd = 10)  
sample_data

 [1] 44.39524 47.69823 65.58708 50.70508 51.29288 67.15065 54.60916 37.34939
 [9] 43.13147 45.54338

Vectorization

Vectorization refers to performing an operation on all items of a vector at once instead of going through each item in a loop.

Code using vectorized functions are also much cleaner than using for loop

Example: Given a vector of numeric, and square all the odd numbers while divide even numbers by 2

non_vectorized <- function(numbs){
  for (i in 1:length(numbs)){ # go through each number in numbs
    if (numbs[i] %%2 == 0){
      numbs[i] <- numbs[i]/2
    }else{
      numbs[i] <- numbs[i]**2
    }
  }
  
  numbs
}

vectorized <- function(numbs){
  numbs <- ifelse(numbs %%2 == 0, numbs/2, numbs**2)
  
  numbs
}

Vectorization

Example: Compare run time between 2 approahces

# non vectorization approach
system.time(
  non_vectorized(c(1:100000))
)

   user  system elapsed 
   0.02    0.02    0.03

# vectorized approach
system.time(
  vectorized(c(1:100000))
)

   user  system elapsed 
   0.01    0.00    0.01

Scope

Discussion Read the code and guess the answer for 3 Questions below

example_var <- 10

example_func <- function(arg){
  print(paste0("Value of variable example_var: ", example_var))
  
  example_var <- 20
  
  print(paste0("Value of variable example_var after assignment: ", example_var))
  print(paste0("Value of function argument: ", arg)) 
}

# Q1: What will the 2 print statements show?
example_func(60) 
# Q2: What will be the value for example_var here?
paste0("Value of variable example_var: ", example_var)
# Q3: What will be the value for arg here?
arg

Scope

Output

example_var <- 10

example_func <- function(arg){
  print(paste0("Value of variable example_var: ", example_var))
  
  example_var <- 20
  
  print(paste0("Value of variable example_var after assignment: ", example_var))
  print(paste0("Value of function argument: ", arg)) 
}

# Q1: What will the 2 print statements show?
example_func(60)

[1] "Value of variable example_var: 10"
[1] "Value of variable example_var after assignment: 20"
[1] "Value of function argument: 60"

# Q2: What will be the value for example_var here?
paste0("Value of variable example_var: ", example_var)

[1] "Value of variable example_var: 10"

# Q3: What will be the value for arg here?
arg

Error: object 'arg' not found

Scope

Explanation

# this is a global variable since it is not created in any local environment
example_var <- 10

example_func <- function(arg){
  # global_var
  print(paste0("Global variable can be accessed within function ", example_var))
  
  # this is creating a local variable named example_var
  # it will not affect the global variable example_var
  example_var <- 20
  
  # this example_var refers to the local variable example_var
  print(paste0("Value of variable example_var after assignment: ", example_var))
  # note that argument is also only accessible within the function
  print(paste0("Function argument: ", arg))
}

example_func(60)

[1] "Global variable can be accessed within function 10"
[1] "Value of variable example_var after assignment: 20"
[1] "Function argument: 60"

# value for argument should only accessible within function 
arg

Error: object 'arg' not found

# refers to global variable example_var here 
paste0("Value of variable example_var: ", example_var)

[1] "Value of variable example_var: 10"

Scope

Scope refers to where variables can be accessed or modified.

To put simply

By default, variables can be modified only in the environment in which it was created (unless they were forced to, but it is not recommended).
global variables can be accessed from both outside and inside the function (i.e. from global or local environment).
local variables are variables defined in a local environment and can only be accessed within that local environment. The most common local environment you will encounter is environment within a function.

Packages

Packages are shareable bundle of functions (and sometimes include built-in dataset). We will frequently use available packages to assist us during coding process.

2 options to install an R packages

Via RStudio interface.
- In the File/Plots/Packages/Help panel
- Click Packages > Install
- Type name of the packages to be installed separated by ,
- Click install
Using R code install.packages("package_name")

Packages

Example: Download package for reading Excel files

install.packages("readxl")

Update renv.lock

To update your renv.lock after changing development environment, run the code renv::snapshot()

Packages

To use a package, you must first load it to R using library() function

library(readxl)
data <- read_excel("../data/linelist_raw.xlsx")

New names:
• `` -> `...28`

head(data)

# A tibble: 6 × 28
  case_id generation `infection date`    `date onset` `hosp date`        
  <chr>        <dbl> <dttm>              <chr>        <dttm>             
1 5fe599           4 2014-05-08 00:00:00 2014-05-13   2014-05-15 00:00:00
2 8689b7           4 NA                  2014-05-13   2014-05-14 00:00:00
3 11f8ea           2 NA                  2014-05-16   2014-05-18 00:00:00
4 b8812a           3 2014-05-04 00:00:00 2014-05-18   2014-05-20 00:00:00
5 893f25           3 2014-05-18 00:00:00 2014-05-21   2014-05-22 00:00:00
6 be99c8           3 2014-05-03 00:00:00 2014-05-22   2014-05-23 00:00:00
# ℹ 23 more variables: date_of_outcome <dttm>, outcome <chr>, gender <chr>,
#   hospital <chr>, lon <dbl>, lat <dbl>, infector <chr>, source <chr>,
#   age <chr>, age_unit <chr>, row_num <dbl>, wt_kg <dbl>, ht_cm <dbl>,
#   ct_blood <dbl>, fever <chr>, chills <chr>, cough <chr>, aches <chr>,
#   vomit <chr>, temp <dbl>, time_admission <chr>, merged_header <chr>,
#   ...28 <chr>

Packages

Similar to functions, you can also look up documentation for packages using ? or help()

# --- look up documentation for package ---
?readxl
help(readxl)

# --- look up documentation for function of the package ---
?readxl::read_excel
help(read_excel, package= readxl)

In-class exercise

Goal

Note

In this exercise, you will:

Create a function from the given code
Push your code to Github
Commenting collaborators’ code on Github

Code file

In your R project, navigate to the template code file day1/inclass.R.

Then run all the code in the file, this should create a dataset covid_cases.

This is also where you will write the solution for in-class exercise.

Create function

Dataset Given a dataset with number of reported cases for 211 countries and archipelagos

Where each column following the format cases_{country_code} represents the number of newly reported Covid cases on the day shown in date columns

── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
✔ dplyr     1.1.4     ✔ readr     2.1.5
✔ forcats   1.0.0     ✔ stringr   1.5.1
✔ ggplot2   3.5.1     ✔ tibble    3.2.1
✔ lubridate 1.9.4     ✔ tidyr     1.3.1
✔ purrr     1.0.2     
── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
✖ dplyr::filter() masks stats::filter()
✖ dplyr::lag()    masks stats::lag()
ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errors

head(covid_cases)

        date cases_chn cases_kor cases_aus cases_jpn cases_mys cases_phl
1 2020-01-20         0         0         0         0         0         0
2 2020-01-21        31         0         0         0         0         0
3 2020-01-23       262         0         0         0         0         0
4 2020-01-24       259         1         0         0         0         0
5 2020-01-25       467         0         3         2         0         0
6 2020-01-26       688         0         1         0         3         0
  cases_sgp cases_nzl cases_vnm cases_brn cases_khm cases_mng cases_fji
1         0         0         0         0         0         0         0
2         0         0         0         0         0         0         0
3         0         0         0         0         0         0         0
4         1         0         2         0         0         0         0
5         2         0         0         0         0         0         0
6         1         0         0         0         0         0         0
  cases_lao cases_png cases_gum cases_pyf cases_ncl cases_mnp cases_idn
1         0         0         0         0         0         0         0
2         0         0         0         0         0         0         0
3         0         0         0         0         0         0         0
4         0         0         0         0         0         0         0
5         0         0         0         0         0         0         0
6         0         0         0         0         0         0         0
  cases_tha cases_ind cases_lka cases_mdv cases_bgd cases_btn cases_npl
1         0         0         0         0         0         0         0
2         0         0         0         0         0         0         0
3         2         0         0         0         0         0         0
4         0         0         0         0         0         0         0
5         0         0         0         0         0         0         1
6         1         0         0         0         0         0         0
  cases_mmr cases_tls cases_usa cases_can cases_bra cases_chl cases_ecu
1         0         0         0         0         0         0         0
2         0         0         0         0         0         0         0
3         0         0         1         0         0         0         0
4         0         0         0         0         0         0         0
5         0         0         1         0         0         0         0
6         0         0         0         0         0         0         0
  cases_mex cases_dom cases_pan cases_per cases_arg cases_col cases_cri
1         0         0         0         0         0         0         0
2         0         0         0         0         0         0         0
3         0         0         0         0         0         0         0
4         0         0         0         0         0         0         0
5         0         0         0         0         0         0         0
6         0         0         0         0         0         0         0
  cases_ury cases_cub cases_hnd cases_ven cases_bol cases_tto cases_pry
1         0         0         0         0         0         0         0
2         0         0         0         0         0         0         0
3         0         0         0         0         0         0         0
4         0         0         0         0         0         0         0
5         0         0         0         0         0         0         0
6         0         0         0         0         0         0         0
  cases_gtm cases_jam cases_slv cases_brb cases_bhm cases_guy cases_hti
1         0         0         0         0         0         0         0
2         0         0         0         0         0         0         0
3         0         0         0         0         0         0         0
4         0         0         0         0         0         0         0
5         0         0         0         0         0         0         0
6         0         0         0         0         0         0         0
  cases_lca cases_dma cases_grd cases_sur cases_kna cases_atg cases_nic
1         0         0         0         0         0         0         0
2         0         0         0         0         0         0         0
3         0         0         0         0         0         0         0
4         0         0         0         0         0         0         0
5         0         0         0         0         0         0         0
6         0         0         0         0         0         0         0
  cases_blz cases_vct cases_pri cases_mtq cases_glp cases_abw cases_guf
1         0         0         0         0         0         0         0
2         0         0         0         0         0         0         0
3         0         0         0         0         0         0         0
4         0         0         0         0         0         0         0
5         0         0         0         0         0         0         0
6         0         0         0         0         0         0         0
  cases_vir cases_bmu cases_cym cases_sxm cases_maf cases_cuw cases_blm
1         0         0         0         0         0         0         0
2         0         0         0         0         0         0         0
3         0         0         0         0         0         0         0
4         0         0         0         0         0         0         0
5         0         0         0         0         0         0         0
6         0         0         0         0         0         0         0
  cases_msr cases_tca cases_aia cases_vgb cases_bes cases_flk cases_spm
1         0         0         0         0         0         0         0
2         0         0         0         0         0         0         0
3         0         0         0         0         0         0         0
4         0         0         0         0         0         0         0
5         0         0         0         0         0         0         0
6         0         0         0         0         0         0         0
  cases_esp cases_ita cases_deu cases_fra cases_gbr cases_tur cases_bel
1         0         0         0         0         0         0         0
2         0         0         0         0         0         0         0
3         0         0         0         0         0         0         0
4         0         0         0         0         0         0         0
5         0         0         0         3         0         0         0
6         0         0         0         0         0         0         0
  cases_che cases_nld cases_prt cases_aut cases_rus cases_isr cases_swe
1         0         0         0         0         0         0         0
2         0         0         0         0         0         0         0
3         0         0         0         0         0         0         0
4         0         0         0         0         0         0         0
5         0         0         0         0         0         0         0
6         0         0         0         0         0         0         0
  cases_irl cases_nor cases_dnk cases_pol cases_cze cases_rou cases_lux
1         0         0         0         0         0         0         0
2         0         0         0         0         0         0         0
3         0         0         0         0         0         0         0
4         0         0         0         0         0         0         0
5         0         0         0         0         0         0         0
6         0         0         0         0         0         0         0
  cases_srb cases_fin cases_ukr cases_grc cases_isl cases_hrv cases_mda
1         0         0         0         0         0         0         0
2         0         0         0         0         0         0         0
3         0         0         0         0         0         0         0
4         0         0         0         0         0         0         0
5         0         0         0         0         0         0         0
6         0         0         0         0         0         0         0
  cases_est cases_hun cases_svn cases_blr cases_ltu cases_arm cases_aze
1         0         0         0         0         0         0         0
2         0         0         0         0         0         0         0
3         0         0         0         0         0         0         0
4         0         0         0         0         0         0         0
5         0         0         0         0         0         0         0
6         0         0         0         0         0         0         0
  cases_bih cases_kaz cases_svk cases_mkd cases_bgr cases_uzb cases_and
1         0         0         0         0         0         0         0
2         0         0         0         0         0         0         0
3         0         0         0         0         0         0         0
4         0         0         0         0         0         0         0
5         0         0         0         0         0         0         0
6         0         0         0         0         0         0         0
  cases_lva cases_cyp cases_alb cases_smr cases_mlt cases_kgz cases_mne
1         0         0         0         0         0         0         0
2         0         0         0         0         0         0         0
3         0         0         0         0         0         0         0
4         0         0         0         0         0         0         0
5         0         0         0         0         0         0         0
6         0         0         0         0         0         0         0
  cases_geo cases_lie cases_mco cases_vat cases_fro cases_ggy cases_jey
1         0         0         0         0         0         0         0
2         0         0         0         0         0         0         0
3         0         0         0         0         0         0         0
4         0         0         0         0         0         0         0
5         0         0         0         0         0         0         0
6         0         0         0         0         0         0         0
  cases_imn cases_gib cases_grl cases_irn cases_pak cases_sau cases_qat
1         0         0         0         0         0         0         0
2         0         0         0         0         0         0         0
3         0         0         0         0         0         0         0
4         0         0         0         0         0         0         0
5         0         0         0         0         0         0         0
6         0         0         0         0         0         0         0
  cases_egy cases_irq cases_are cases_mar cases_bhr cases_lbn cases_tun
1         0         0         0         0         0         0         0
2         0         0         0         0         0         0         0
3         0         0         0         0         0         0         0
4         0         0         0         0         0         0         0
5         0         0         0         0         0         0         0
6         0         0         0         0         0         0         0
  cases_jor cases_kwt cases_omn cases_afg cases_dji cases_syr cases_lby
1         0         0         0         0         0         0         0
2         0         0         0         0         0         0         0
3         0         0         0         0         0         0         0
4         0         0         0         0         0         0         0
5         0         0         0         0         0         0         0
6         0         0         0         0         0         0         0
  cases_sdn cases_som cases_yem cases_pse cases_zaf cases_dza cases_bfa
1         0         0         0         0         0         0         0
2         0         0         0         0         0         0         0
3         0         0         0         0         0         0         0
4         0         0         0         0         0         0         0
5         0         0         0         0         0         0         0
6         0         0         0         0         0         0         0
  cases_civ cases_sen cases_gha cases_cmr cases_nga cases_mus cases_cod
1         0         0         0         0         0         0         0
2         0         0         0         0         0         0         0
3         0         0         0         0         0         0         0
4         0         0         0         0         0         0         0
5         0         0         0         0         0         0         0
6         0         0         0         0         0         0         0
  cases_rwa cases_mdg cases_ken cases_zmb cases_tgo cases_uga cases_eth
1         0         0         0         0         0         0         0
2         0         0         0         0         0         0         0
3         0         0         0         0         0         0         0
4         0         0         0         0         0         0         0
5         0         0         0         0         0         0         0
6         0         0         0         0         0         0         0
  cases_ner cases_cog cases_tza cases_mli cases_gin cases_gnq cases_nam
1         0         0         0         0         0         0         0
2         0         0         0         0         0         0         0
3         0         0         0         0         0         0         0
4         0         0         0         0         0         0         0
5         0         0         0         0         0         0         0
6         0         0         0         0         0         0         0
  cases_swz cases_moz cases_syc cases_gab cases_ben cases_caf cases_eri
1         0         0         0         0         0         0         0
2         0         0         0         0         0         0         0
3         0         0         0         0         0         0         0
4         0         0         0         0         0         0         0
5         0         0         0         0         0         0         0
6         0         0         0         0         0         0         0
  cases_cpv cases_tcd cases_mrt cases_zwe cases_gmb cases_lbr cases_ago
1         0         0         0         0         0         0         0
2         0         0         0         0         0         0         0
3         0         0         0         0         0         0         0
4         0         0         0         0         0         0         0
5         0         0         0         0         0         0         0
6         0         0         0         0         0         0         0
  cases_gnb cases_bwa cases_mwi cases_stp cases_bdi cases_sle cases_ssd
1         0         0         0         0         0         0         0
2         0         0         0         0         0         0         0
3         0         0         0         0         0         0         0
4         0         0         0         0         0         0         0
5         0         0         0         0         0         0         0
6         0         0         0         0         0         0         0
  cases_reu cases_myt
1         0         0
2         0         0
3         0         0
4         0         0
5         0         0
6         0         0

Create function

Goal Create a function that allow users to choose compute basic statistics (min, max, mean, quartiles) for number cases of each country.

Given the code to compute basic stats for case reports in China (cases_chn) is as followed

case_col <- "cases_chn"

basic_stats <- c(
  min =  min(covid_cases[[case_col]]),
  q1 = quantile(covid_cases[[case_col]], 0.25, names=FALSE),
  median = median(covid_cases[[case_col]]),
  q3 = quantile(covid_cases[[case_col]], 0.75, names=FALSE),
  max = max(covid_cases[[case_col]])
)

basic_stats

    min      q1  median      q3     max 
    0.0    65.5   128.0   950.5 19461.0

Note

The syntax data[[column_name]] is used for extracting values in a column

Navigating Github Desktop & Github

Once you have created the function, complete the following tasks on Github Desktop and Github (refer to the handout for detailed instructions)

Note

Push your code to Github using Github Desktop
Create an issue on Github to comment on at least 1 other participant’s code

Data manipulation in R

Data import

Some useful import functions

Function	Package	Description
`read_excel()`	`readxl`	import Excel (`.xlsx`, `.xls`) files
`read_csv()`	`readr`	import `.csv` files
`readRDS()`	`base`	import `.rds` files
`load()`	`base`	import `.rda` files

Data import

rio - support data import for multiple data format

library(rio)
data <- import("../data/linelist_raw.xlsx")

New names:
• `` -> `...28`

Data export

Some useful export function in R

Function	Package	Description
`write_xlsx()`	`writexl`	export data as Excel files
`write.csv()`	`utils`	export data as `.csv` files
`saveRDS()`	`base`	export data as `.rds` files
`save()`	`base`	export data `.rda` files

Clean column names

The thing we usually do first after importing data is check the column names, to do that we use function colnames()

colnames(data)

 [1] "case_id"         "generation"      "infection date"  "date onset"     
 [5] "hosp date"       "date_of_outcome" "outcome"         "gender"         
 [9] "hospital"        "lon"             "lat"             "infector"       
[13] "source"          "age"             "age_unit"        "row_num"        
[17] "wt_kg"           "ht_cm"           "ct_blood"        "fever"          
[21] "chills"          "cough"           "aches"           "vomit"          
[25] "temp"            "time_admission"  "merged_header"   "...28"

Clean column names

You may notice some names go against the “rule of thumb” for naming

To quickly resolve this, package janitor offers clean_names() function to automatically fix the naming

library(janitor)


Attaching package: 'janitor'

The following objects are masked from 'package:stats':

    chisq.test, fisher.test

data <- clean_names(data)
head(data)

  case_id generation infection_date date_onset  hosp_date date_of_outcome
1  5fe599          4     2014-05-08 2014-05-13 2014-05-15            <NA>
2  8689b7          4           <NA> 2014-05-13 2014-05-14      2014-05-18
3  11f8ea          2           <NA> 2014-05-16 2014-05-18      2014-05-30
4  b8812a          3     2014-05-04 2014-05-18 2014-05-20            <NA>
5  893f25          3     2014-05-18 2014-05-21 2014-05-22      2014-05-29
6  be99c8          3     2014-05-03 2014-05-22 2014-05-23      2014-05-24
  outcome gender                             hospital       lon      lat
1    <NA>      m                                Other -13.21574 8.468973
2 Recover      f                                 <NA> -13.21523 8.451719
3 Recover      m St. Mark's Maternity Hospital (SMMH) -13.21291 8.464817
4    <NA>      f                        Port Hospital -13.23637 8.475476
5 Recover      m                    Military Hospital -13.22286 8.460824
6 Recover      f                        Port Hospital -13.22263 8.461831
  infector source age age_unit row_num wt_kg ht_cm ct_blood fever chills cough
1   f547d6  other   2    years       1    27    48       22    no     no   yes
2     <NA>   <NA>   3    years       2    25    59       22  <NA>   <NA>  <NA>
3     <NA>   <NA>  56    years       3    91   238       21  <NA>   <NA>  <NA>
4   f90f5f  other  18    years       4    41   135       23    no     no    no
5   11f8ea  other   3    years       5    36    71       23    no     no   yes
6   aec8ec  other  16    years       6    56   116       21    no     no   yes
  aches vomit temp time_admission merged_header x28
1    no   yes 36.8           <NA>             a   b
2  <NA>  <NA> 36.9          09:36             a   b
3  <NA>  <NA> 36.9          16:48             a   b
4    no    no 36.8          11:22             a   b
5    no   yes 36.9          12:60             a   b
6    no   yes 37.6          14:13             a   b