diffExApp

An interactive application for differential expression analysis.

This app takes a single file (.csv, .tsv, or .txt). The file should contain columns of raw gene counts data with a single row of Gene ID's (ID can be in any form). Once running the user can select replicates and run a simple pairwise gene expression analysis using either EdgeR or DESeq2.

Curious what exactly is happening under the hood? Check out the differential expression code here. The functions edgerDE() and deseqDE() handle the bulk of the analysis.

Installation

The easiest way to install this application is to clone it from this GitHub. Open the command line (terminal on Mac) and type the following commands:

Go to the directory that you want to download the app to.

cd <PATH/TO/DIR>

Clone this repository

git clone https://github.com/FredHutch/interactiveVolcano.git

Enter the repository and switch to the local branch

cd interactiveVolcano

Set up

• This app is set up to run in a specific Docker environment. Most likely you'll be running this application locally so you'll have to make a quick change to app.R. To switch the application to local, open app.R and switch local <- TRUE.
• You can now run the app with it's example dataset. A .csv derived from the airway dataset.
  • To run the app either click run app in the right corner of the RStudio source pane when you have app.R open or by running R -e "shiny::runApp('~/path/to/shinyapp')" on the command line.
• Once you've checked that the app runs with the example data, use you're own dataset by replacing the example data with any counts data of your own!
  • If you want a place to archive data files you can create a directory called archive in data/ and keep data files there without disrupting the app.

Usage

Once the application is running, it opens to a tab where you can select replicates and decide on an analysis package (either edgeR or DESeq2). Once everything is set, click the apply button and the differential expression analysis will run. You'll see a PCA plot appear underneath the sample table when your results are ready.

The results are on the next tab. On the left hand sidebar, there are thresholds for significance and effect size to determine what is considered differentially expressed. You can view the differential expression results table, an MA plot, a volcano plot, and a clustered heatmap of differentially expressed genes by clicking the tabs on the right hand side. Below each results table or plot there are options for further customization and a download button for that specific result.

What to expect going forward

Currently this app runs a very simplified gene expression analysis. We're working to bring more complex analyses to this shiny app, stay tuned!