setup.Rmd

---
title: "Setup"
output: html_document
---

### R

**Install R.** You'll need R version **4.2** or higher.[^rversion] Download and install R for [Windows](http://cran.r-project.org/bin/windows/base/) or [Mac](http://cran.r-project.org/bin/macosx/) (download the latest R-4.x.x.pkg file for your appropriate version of Mac OS).

### RStudio

Download and install **[RStudio Desktop](https://www.rstudio.com/products/rstudio/download/)**.

R and RStudio are separate downloads and installations. **R** is the underlying statistical computing environment. **R** is quite usable by itself, but **RStudio** is a graphical integrated development environment that makes using **R** for data analysis a bit more efficient. *You need R installed before you install RStudio*.

### CRAN packages

We will need to install packages throughout the course. To get started, launch **RStudio** (RStudio, *not R itself*). *Internet access is required for installing packages*. Copy and paste the following commands, one-at-a-time, into the **Console** panel (usually the lower-left panel, by default) and hit the Enter/Return key. If you receive an error message when trying to install any particular package, we can work together to determine a fix.

```r
install.packages("dplyr")
install.packages("readr")
install.packages("tidyr")
install.packages("ggplot2")
```

_A few notes_:

- Commands are case-sensitive. 
- You must be connected to the internet.
- Even if you've installed these packages in the past, do re-install the most recent version. Many of these packages are updated often, and we may use new features in the workshop that aren't available in older versions.
- If you're using Windows you might see errors about not having permission to modify the existing libraries -- disregard these. You can avoid this by running RStudio as an administrator (right click the RStudio icon, then click "Run as Administrator").
- These core packages are part of the "tidyverse" ecosystem (see [tidyverse.org](https://www.tidyverse.org/)). There is a [tidyverse](http://tidyverse.org/) package that's kind of a meta-package that automatically installs/loads all of the above packages and several other commonly used packages for data analysis that all play well together.[^tidyverse] You could optionally install the tidyverse package instead of all these packages individually. See [tidyverse.org](https://www.tidyverse.org/) for more.

[^rversion]: If you have not updated your R installation since then, you should upgrade to a more recent version, since several of the required packages depend on a version at least this recent. You can check your R version by typing `version` at the R prompt.

[^tidyverse]: Installing/loading the tidyverse **tidyverse** will install/load the core tidyverse packages that you are likely to use in almost every analysis:
**ggplot2** (for data visualisation), **dplyr** (for data manipulation), **tidyr** (for data tidying), **readr** (for data import), **purrr** (for functional programming), and **tibble** (for tibbles, a modern re-imagining of data frames). It also installs a selection of other tidyverse packages that you're likely to use frequently, but probably not in every analysis (these are installed, but you'll have to load them separately with `library(packageName)`). This includes: **hms** (for times), **stringr** (for strings), **lubridate** (for date/times), **forcats** (for factors), **DBI** (for databases), **haven** (for SPSS, SAS and Stata files), **httr** (for web apis), **jsonlite** (or JSON), **readxl** (for .xls and .xlsx files), **rvest** (for web scraping), **xml2** (for XML), **modelr** (for modelling within a pipeline), and **broom** (for turning models into tidy data). After installing tidyverse with `install.packages("tidyverse")` and loading it with `library(tidyverse)`, you can use `tidyverse_update()` to update all the tidyverse packages installed on your system at once.

### Bioconductor

Bioconductor uses its own installation procedure. The CRAN package, BiocManager, is the key to installing
Bioconductor packages. Start by installing `BiocManager`.

```r
install.packages("BiocManager")
```

To install basic Bioconductor packages (we will be installing more throughout the course), type the following:


```r
BiocManager::install()
```

Installing a specific Bioconductor package works similarly to `install.packages` and you can use `BiocManager::install()` as a replacement for `install.packages`.

```r
BiocManager::install("GEOquery")
```


Consider using the `BiocManager::install()` installation process for all other R packages. BiocManager
can install from CRAN, Bioconductor, and several other repositories including GitHub.


### Check that things worked

Check that you've installed everything correctly by closing and reopening RStudio and entering the following command at the console window (don't worry about any messages that look something like `the following objects are masked from ...`[^masking], or `Warning message: package ... was build under R version ...`[^oldversion]):

```r
library(dplyr)
library(readr)
library(tidyr)
library(ggplot2)
```

[^masking]: We'll talk about this in class. It's not a concern.

[^oldversion]: This means the version of R you have installed is older than the version that the package author used when they built the package you're trying to use. 99% of the time it isn't a problem, unless your R version is very old (you should be using 3.5.0 or later for this course).

This may produce some notes or other output, but as long as you don't get an error message, you're good to go. If you get a message that says something like: `Error in library(somePackageName) : there is no package called 'somePackageName'`, then the required packages did not install correctly. Either try reinstalling or contact the instructor.