Learn to R blog series - R and RStudio

Hello everyone, welcome back! This post marks the beginning, hopefully, of your foray into the wonderful world of R and RStudio…and my delve into the odd vlog to go with the blog! I’ve brushed my hair for you, and I don’t do that for just anyone, so you’d better watch it at least once! So without further ado.

R

R is an open source language released in 2001 that’s ideal for data wrangling and data science. It has connectors to pretty every much every data source under the sun, allows you wrangle data like nobody’s business, build pretty much every type of model ever thought up, and visualise it in all the niftiest ways. These superlatives are not disingenuous, R really is that broad and amazing.

R is a vibrant ecosystem that enables people to extend, enhance, and replace any part of it. There are many paradigms in R to facilitate object-oriented programming, functional programming, and more. If you can write something in C++, FORTRAN, Python, or JavaScript–and of course, R!–you can write extensions for R.

There are currently more than eleven thousand extensions (referred to as packages) to R in the core ecosystem (which is a fancy word for the collected bits and pieces of R!) and two and a half thousand packages in the genomics ecosystem.

We’re also seeing emerging ecosystems and paradigms within CRAN. The tidyverse is one such ecosystem, focussed primarily on analysing tabular data, and it will be used in future works extensively.

CRAN

The core ecosystem is CRAN, the Comprehensive R Archive Network. CRAN is maintained by some great people who put in place a large number of quality gates that an R package must adhere to in order to be made widely available. They then host these packages and do great things like daily re-runs of all package tests to ensure packages are still working. CRAN is the default source of packages for most R users.

If you use RStudio, you’ll use a mirror of CRAN hosted by RStudio. There are a number of these mirrors scattered over the globe to help reduce the load on the central servers. You can use another one of these mirrors, or even set-up your own internal CRAN.

Key points to know about R

  • R works in-memory which means that the processing is fast but the amount of data you can process is limited to how much RAM your data takes up and how much your computations will require.
  • R is not multi-threaded by default, it works on a single CPU core. Making use of more than one of your cores to spread the load requires additional packages and often additional coding.
  • R is quirky! In some ways, R is a lot like other common programming languages, which can make it pretty easy to pick up. However, because R is still designed to be compatible with S, it’s actually pretty darn old and as a result, really odd in places.
  • Coding R will give you the typical gotcha’s, and add another: case sensitivity. R is (un)fortunately a language where “Red” and “red” are different and this also extends to variable and function names (which we’ll discuss later.) As a consequence, the most common errors you’ll find when writing code in R are:
    • Incorrectly placed or missing commas
    • Incorrectly placed or missing brackets
    • Incorrectly placed or missing operators
    • Incorrect case used when typing
  • With so many packages available to extend R, the answer to “how do I write this?” is usually “there’s more than one package for that”.

Summary

R is an great language for doing data analysis, data science, and more. It has it’s quirks but the community around it is huge and is making R easier to adopt every day.

Why use R?

R as a programming language is brilliant at its core competencies – statistics and data visualisation. It’s also a great “glue” language, by which I mean that you can use it to perform computations in many different languages and combine the results smoothly. As a result, R enables you to be an effective data wrangler, data scientist, and/or data visualisation practitioner.

So, now we’ve done the hard sell, here’s some fun stuff… and you can’t even acuse us of being on commision…because R and RStudio are totally free (yes free) to download and use what with them being opensource and whatnot.

Video

Now it’s time for this weeks show and tell! I know you’ve brought a super cool bug thing in a tupperware for us all to look at, but I brought RStudio, so I get to go first.

Search