Skip to content

Fiji with MoBIE

Introduction

Fiji is an image processing tool that combines the ImageJ software with community plugins for scientific image processing and analysis. The RFI fiji-mobie container includes the MoBIE (Multimodal Big Image Data Exploration) plugin designed for sharing and interactively browsing large, multi-modal images and associated tabular data.

Using Fiji and MoBIE

Prerequisites: Fiji is a graphical application that requires a local or remote display server such as X11 to render its interface. It does not require a GPU.

The Fiji application can be started by pulling and running the container using Apptainer: 

apptainer run docker://quay.io/rosalindfranklininstitute/fiji-mobie
It may also be run using Docker. First, execute xhost +local: (or xhost + when using a remote system) to allow containers to access the current display, then run: 
docker run \
  -v /path/to/your/data:/data \
  -e DISPLAY=$DISPLAY \
  -v /tmp/.X11-unix:/tmp/.X11-unix \
  --device=/dev/dri:/dev/dri \
  quay.io/rosalindfranklininstitute/fiji-mobie

In each case, if you want to access files outside of the default directories made available by Apptainer or Docker, add --bind /path/to/files:/mnt (Apptainer) or -v /path/to/files:/mnt (Docker). Replace /path/to/files with a location on your machine; /mnt is where those files will appear inside the container (for example, when using the Fiji file picker File>Open...).

Warning

When you start Fiji, a pop-up may prompt you to run the Fiji Updater. Close this or select No/Never—even if the update was successful, it requires a restart to apply, and so would be lost next time you started the container since the software does not persist between container sessions.

Loading image data

The Fiji application interface consists of a small pane that expands when an image is loaded.

The Fiji GUI The Fiji application toolbar. Status text provides information on the currently selected option, and a search bar may be used to easily find commands.

Fiji supports a wide range of image formats (see here for a list) including .tiff, .jpeg and .bmp. You can also load and save stacks as image sequences. Below is the Fiji image pane displaying Fluorescence data (Fluoro 01.tif) from the Monash University Demo Images1.

The Fiji Image Pane
Fluoroscene image data in Fiji. Fiji can be used to manipulate images interactively, such as splitting the red (blood vessels) and green (hypoxia) channels.

The Fiji website has free tutorials, manuals and presentations if you want to learn more about the range of image processing abilities the application has. It also explains how you can contribute to its development.

Running MoBIE

MoBIE uses the BigDataViewer to provide smooth browsing of large image datasets. This includes support for next-gen image formats such as OME-Zarr and HDF5. You can perform image registration directly in MoBIE views and explore associated tabular data alongside images as you view them.

While BigDataViewer can load view local images, MoBIE adds support for loading cloud-based projects. To access these, enter "mobie" in the Fiji search bar and select Open MoBIE Project. Pressing Run will prompt you to enter a GitHub repository containing the target datasets. An example dataset is available at https://github.com/mobie/platybrowser-project2. After entering the repository URL, the MoBIE project pane opens alongside the BigDataViewer, with the layered image data ready for you to inspect.

MoBIE Open Project Command MoBIE Example Dataset Loading image data from a repository in MoBIE. The mouse scroll wheel can be used to traverse the layer images. Images show muscle in an Epithelial cell from Fig. 1D of the Cell publication2. Other views can be selected via the cell-paper drop-down menu .

The MoBIE Tutorials Site has examples and tutorials for learning more about using MoBIE to perform image registration and annotation. We recommend starting with the Basic image browsing tutorial to understand both the BigDataViewer interface and the structure of Datasets that MoBIE reads.

Troubleshooting and Support

Getting help

If you encounter problems with Fiji or MoBIE, you can Create an issue on the container GitHub repository. Please share as much information as you can regarding your host system (operating system, container runtime), the commands you are trying to run as well as any error messages.

Version and license information

The container currently has

  • Fiji 2.16.1 (ImageJ2 2.16.0/1.54p)
  • MoBIE 4.0.3

The container is distributed under an Apache 2.0 License. If you are using this software in your work, consider giving attribution to the Fiji3 and MoBIE4 publications.

  1. Cameron Nowell, Fiji Manual Resources v6.5 (2023). ↩

  2. Vergara, H. M. et al., Cell 184 (18), 4819-4837.e22. ↩↩

  3. Schindelin, J. et al., Nature Methods, 9, 676–682 (2012). ↩

  4. Pape, C. et al., Nature Methods 20, 475-476 (2023). ↩