PathviewViz

A Shiny app for visualizing KEGG pathways colored by your differential expression results. Upload a DEG table, pick a pathway, and see the genes painted by their fold change, with optional GSEA to help you decide which pathways are worth looking at.

Demo

demo.mp4

What it does

1. Upload a DEG table (CSV or TSV) that has a gene-ID column and a numeric column such as log2FC.
2. Choose the organism, the gene-ID type (SYMBOL, ENSEMBL, ENTREZID, and so on), and which two columns to use.
3. Enter one or more KEGG pathway IDs, just the digits, for example 04110 for the cell cycle. Separate several with commas.
4. Render. The KEGG map appears in the app colored by your values, and you can download it as a PNG.

You can also set the colors for up- and down-regulated genes, and run GSEA (gene set enrichment analysis) to get a ranked list of pathway names to explore. Behind the scenes, gene IDs are mapped to Entrez through the organism's org.*.eg.db package and then handed to pathview.

Running it on your computer

The app runs locally in R; a browser tab opens automatically when it starts.

1. Install R

Download and install R from cran.r-project.org. RStudio Desktop is optional but makes launching the app a one-click affair.

2. Get the code

Clone the repository, or download it as a ZIP from GitHub and unzip it:

git clone https://github.com/<your-username>/PathviewViz.git
cd PathviewViz

3. Install the dependencies (first time only)

From an R session started in the project folder, run:

source("install.R")

This installs the CRAN and Bioconductor packages the app needs (pathview, clusterProfiler, the annotation databases, and so on). It is a sizable download and can take several minutes the first time. The exact package list is also spelled out in install.R if you prefer to install things by hand.

4. Launch the app

Either open app.R in RStudio and click Run App, or start it from an R session:

shiny::runApp()      # run from inside the project folder

A browser window opens with the app. To stop it, press Esc in RStudio or Ctrl-C in the console.

5. Try it with the sample data

A small human example, sample_deg.csv (gene symbols with log2 fold changes), is included. Upload it, keep the defaults (Human, SYMBOL), and try pathway 04110 (cell cycle) or 04115 (p53 signaling).

Notes

• The first time you render a given pathway, the app downloads its data from KEGG and caches it for the rest of the session, so an internet connection is needed.
• KEGG pathway IDs are the five-digit numbers; the organism prefix (hsa, mmu, and so on) is added for you.
• Each session works in its own temporary folder, so several people can use a shared deployment without overwriting each other's images.