Code repository for "Differential co-localisation analysis of multi-sample and multi-condition experiments with spatialFDA"

Repository organising the code to generate the results of the spatialFDA manuscript. spatialFDA is available on Bioconductor

The analysis is written as a snakemake workflow. In order to run the scripts install snakemake

Requirements

Dependencies

To run this project you need to have installed snakemake --version 9.16.3, conda or a similar option like mamba and apptainer. You can also run this project with conda only, but then you have to specify your python environments yourself.

Data

The dataset is available through Bioconductors ExperimentHub via the package imcdatasets and downloaded directly in the workflow. The original data was deposited at https://data.mendeley.com/datasets/cydmwsfztj/2.

Quick start

In order to run the self-contained reproducible environment please run:

1. Clone analysis repo and move to the correct directory

git clone https://github.com/mjemons/spatialfda-manuscript.git
cd spatialfda-manuscript/

2. Run the workflow

snakemake --cores <nCores> --sdm conda apptainer

If you want to use only the conda environment without apptainer you might need to specify a correct python installation.

It can happen that you run into an apptainer error when setting up the environment. In this case please specify a new tmp directory via --apptainer-args

Citation

@article {Emons2026.04.13.718197,
	author = {Emons, Martin and Scheipl, Fabian and Gunz, Samuel and Purdom, Elizabeth and Robinson, Mark D},
	title = {Differential co-localisation analysis of multi-sample and multi-condition experiments with spatialFDA},
	year = {2026},
	doi = {10.64898/2026.04.13.718197},
	publisher = {Cold Spring Harbor Laboratory},
	journal = {bioRxiv}
}

Contact

For questions, please open an issue in this repository