R packages

code
analysis
plot
Author

Sangbaptist

Published

October 1, 2023

What is a package in R language?

An R package is a collection of functions, data and compiled code that is bundled together in a standardized way to extend the functionality of R. In R, packages allow users to document, share and reuse codes.

I would recommend these books on R packages: Hadley Wickham & Jennifer Bryian and Hadley Wickham.

Examples packages in R are:

  • tidyverse: This is a collection of some R packages designed to make data analysis easier, efficient and more intuitive.
  • caret: Short for Classification and REgression Training provides uniform interfaces for creating predictive models in R, with tools for data splitting, pre-processing, feature selection, model tuning and many more.
  • ggplot2: It is primarily for data visualization, offering a flexible and structured approach to creating static and dynamic plots.Rnews
  • shiny: Shiny is designed for building interactive web applications that allow users to interact with data in real-time, without the need for advanced web programming knowledge.

Key features of an R package:

  1. Function: In an R package, a function is a piece of code that performs a specific task. Functions are the main building blocks of the package, allowing users to perform calculations, transformations, visualizations, or analyses. These functions reside in the R/ directory and are what users call directly when they use the package. For example, if a package provides a function plot_data(), the function would be defined here. Functions are stored in .R files within the R/ directory but their usage and behavior are also documented in the man/ directory (via the .Rd files).

  2. Data: Many R packages include datasets, especially if they are intended for teaching, demonstration, or testing. These datasets are stored in the data/ directory and are often provided as .rda or .csv files. The data/ directory holds these datasets, while the man/ directory will include documentation on how to use them. The vignettes/ directory may also contain tutorials demonstrating how to work with the datasets.

  3. Documentation: Documentation in an R package describes the functions, data, and overall package usage. It includes user-friendly descriptions, examples, and details about the parameters, input/output, and purpose of each function or dataset. This documentation is stored in the man/ directory as .Rd files and can be accessed by users using ?function_name or help() in R. Documentation is created using .Rd files in the man/ directory, and long-form tutorials or explanations can also be found in the vignettes/ directory.

Why use R package?

While R packages can greatly enhance productivity, simplify complex tasks, and provide access to advanced techniques, they are not always necessary for every task. The decision to use an R package should depend on the task’s complexity, performance needs, and your familiarity with the package. R packages extend the base capabilities of R by providing additional functions, tools, and datasets that are not included in the standard R installation. This allows users to perform more advanced tasks, such as machine learning, data visualization, or web development, without needing to reinvent the wheel.

Here are some few importance or advantages of using R packages:

  1. Reusability: Packages in R provide ready-made solutions that save you from writing code from scratch.
  2. Community Support. There are many R communities that you can get support from including networking, learning opportunities and collaboration, at all level. Examples are CRAN, R Consortium, Github and many more.
  3. Specialised tools: Packages often have specific focus area like genomics, machine learning, finance or visualisation, providing tools tailored for those domains.
  4. Efficiency: Some R packages are built with highly optimized algorithms, often implemented in C, C++, or Fortran, which are faster than pure R code. This makes certain packages more efficient, especially for large datasets or complex operations.

Sources of R packages

When R is installed on your computer, it comes with a set of base packages(e.g. utils, graphics, datasets, etc) and recommended packages (MASS, survival, rpart, etc) by default. The number of these packages (combined) depends on the R version your are using. These packages are essential for performing basic tasks and provide foundational functions for R. However, there are many other available packages from online.

R packages come from several primary sources, each contributing to the growth and accessibility of the R ecosystem.

Two of the biggest sources of R packages are CRAN (Comprehensive R Archive Network) and Bioconductor(an open source project for building tools to analyse genomic data) repositories. For more about repositories, check my other post.

Common package listing in R

How to install a package in R:

Code 
install.packages("package_name") #write the name of the package you
                                  #are intending to install (e.g. rio)

How to list packages installed with R:

Code 
installed.packages() #Do not mistake "installed" with "install"

List currently loaded packages

Code 
.packages()

List all installed packages

Code 
.packages(all.available = TRUE)

Listing available packages in R

Code 
library()

Loading an installed package into R

Code 
library(package_name) #Loading a single package
library(c(package_1, package_2)) # Loading two or more packages
require(package_name) #another way to load a package

Listing packages loaded by default

Code 
getOption("defaultPackages")

References:

  1. R Packages by Hadley Wickham & Jennifer Bryian.
  2. Advanced R
  3. R Documentation
  4. Comprehensive R Archive Network(CRAN)
No matching items