GranatumX: A community engaging and flexible software environment for single-cell analysis

Paper location:

Code Authors: GranatumX (alias to current codebase membership located here)

Cloud location

A cloud deployment of GranatumX or GX is located at If you receive an error, try again in 5 minutes as the system may be updating. Also you may use the backup service in the meantime located here in case maintenance is being done on the server. Feel free to run your data through this system, but note that files may be removed after a period of inactivity. You may refer back to this page as a guide to process your data. If you are a tool developer, we include instructions for building your own tools that can be used directly by biologists to process their data and a citation for your work will be given to them.

Video tutorial: using GranatumX

Chinese-accessible video available here.

Getting started using the web interface

Browser compatibility

GX works with most browsers; however, it is optimized for Google Chrome or Safari-based browsers.

Matrix preparation for data

You can upload expression data using a csv, tsv, or zip file. For large matrices, it is recommended to use zip compression on your csv data. The header row should consist of the cell titles. These cell titles will be corresponded to associated data in the metafile data. The column of your data matrix is the gene id. You can typically export the matrix from Excel if that is where your data is located. If you have a 100000 cell table, then your size can be quite large. You should upload your data as a zipped csv file. Note that spacing is ignored.

        ""      ,"cell_1","cell_2","cell_3"
        "gene 1",       0,      10,       2
        "gene 2",       2,       8,       1

Large files require zipping or gzipping (gz). Your metadata should use the same format.

You can upload metadata as a csv, tsv, or zip file as well. This file associates the cell label from the matrix data file to additional data such as cell annotations.

        ""      ,"annotation_1","annotation_2"
        "cell_1", "lung tissue",   "cancerous"
        "cell_2","heart tissue",   "cancerous"
        "cell_3", "lung tissue",      "normal"

You can load the comprehensive recipe and use the example data to test.

If you get stuck at a step or the ui is unresponsive you can reload the page without losing your information.

Running locally

As a prerequisite you will want sufficient RAM (16+ GB ideally), a working installation of docker, and a working bash shell.

For setting up WSL2 on Windows, you can also view the video located here:

Chinese-accessible material available here

For setting up Docker on Windows, you can also view the video located here:

Chinese-accessible material available here

For getting GranatumX up and running you will want to execute in your shell:

        source <( docker run --rm -it granatumx/scripts:1.0.0 )

If you receive an error, run:

        curl -s | source /dev/stdin

Creating your own GBOX

After you have a Bash terminal installed and a copy of Docker working, you can follow the steps below and watch the video below.

Chinese-accessible video available here.