Setup – Inference for High-dimensional Data

Installation

R is a programming language that is especially powerful for data exploration, visualization, and statistical analysis. To interact with R, we use RStudio.

Install the latest version of R from CRAN. If you are using a JAX-owned machine, you can use the JAX Self Service app instead without needing support from the IT help desk.
Install the latest version of RStudio here. Choose the free RStudio Desktop version for Windows, Mac, or Linux. If you are using a JAX-owned machine, you can use the JAX Self Service app instead without needing support from the IT help desk.
Start RStudio. We will use several packages from CRAN. You can install them from the Console or from the Install button on the RStudio Packages tab. Copy-paste this list of packages into the Install dialog box:

devtools, BiocManager, here, rafalib, lasso2, matrixStats

Alternatively, run the following in the Console.
```
 install.packages(c("devtools", "BiocManager", "here", "rafalib", "lasso2", "matrixStats"))
```

Once you have installed the packages, load the libraries by checking the box next to each package name on the Packages tab, or alternatively running this code in the Console for each package.
```
library(devtools)
```
Repeat the command above with the other packages you just installed.
We will also use packages available only on Github. Once the devtools library is loaded you will be able to install packages from Github using install_github(). You must install these packages from the Console. They will not be available from the RStudio Packages tab.
```
 install_github(c("genomicsclass/GSE5859Subset", "genomicsclass/GSE5859", "genomicsclass/maPooling", "genomicsclass/tissuesGeneExpression"))
```
Once you have installed the packages, load the libraries by running this code in the Console.
```
 library(GSE5859Subset)
```
Repeat for each of the packages installed from Github.

Finally, we will use Bioconductor packages. Install these packages in the Console:

 BiocManager::install(c("genefilter", "SpikeInSubset", "SummarizedExperiment", "parathyroidSE", "Biobase", "limma", "qvalue", "PCAtools"))

Once you have installed the Bioconductor packages, load the libraries by running this code in the Console.
```
 library(genefilter)
```

Repeat for each of the packages installed from Bioconductor.

Data files and project organization

Make a new folder in your Desktop called inference. Move into this new folder.
Create a data folder to hold the data, a scripts folder to house your scripts, and a results folder to hold results.

Alternatively, you can use the R console to run the following commands for steps 1 and 2.
```
 setwd("~/Desktop")
 dir.create("./inference")
 setwd("~/Desktop/inference")
 dir.create("./data")
 dir.create("./scripts")
 dir.create("./results")
```
Please download the following file and place it in your data folder. You can download the file from the URL below and move the file the same way that you would for downloading and moving any other kind of data.

Alternatively, you can copy and paste the following into the R console to download the data.

download.file(url = "https://raw.githubusercontent.com/genomicsclass/dagdata/master/inst/extdata/femaleControlsPopulation.csv", destfile = "data/femaleControlsPopulation.csv")

Inference for High-dimensional Data: Setup

Installation

Data files and project organization