This repository contains code for the paper:

Eliason & Rao (2024)
Investigating Ecological Interactions in the Tumor Microenvironment using Joint Species Distribution Models for Point Patterns

We present a statistically rigorous framework for analyzing spatial cell–cell interactions in the tumor microenvironment (TME) using multitype Gibbs point process (MGPP) models. These models are implemented as joint species distribution models (JSDMs) to quantify spatial attraction and repulsion among annotated cell types in multiplexed colorectal cancer images.

This repository enables full reproducibility of the analyses and figures in the manuscript. It includes:

• Preprocessed CODEX multiplexed imaging data for 35 colorectal cancer patients.
• MGPP model fitting functions using a fork of the ppjsdm R package.
• An annotated Quarto notebook reproducing all figures and tables.
• An HPC-compatible batch script for parallel model fitting across patients.

• R version ≥ 4.2
• Packages: tidyverse, spatstat.core, future, future.apply, remotes, quarto

Clone the repository and install required packages:

git clone https://github.com/jeliason/mgpp_TME.git
cd mgpp_TME
# Install required R packages
Rscript -e "install.packages(c('remotes', 'quarto', 'tidyverse', 'spatstat.core', 'future', 'future.apply'))"
Rscript -e "remotes::install_github('jeliason/ppjsdm')"

To reproduce the full example workflow and regenerate all manuscript figures and tables:

quarto::quarto_render("mgpp_example_workflow.qmd")

The HTML output will be saved in _output/.

To run the full MGPP model across all patients in parallel:

1. Modify hpc.R for your computing environment (paths).
2. Submit the job on Slurm using rslurm:

Rscript hpc.R

Please cite our paper if you use this code or data:

@article{eliason2024,
  title = {Investigating {{Ecological Interactions}} in the {{Tumor Microenvironment Using Joint Species Distribution Models}} for {{Point Patterns}}},
  author = {Eliason, Joel and Rao, Arvind},
  year = {2024},
  month = jun,
  journal = {The New England Journal of Statistics in Data Science},
  volume = {2},
  number = {3},
  pages = {296--310},
  publisher = {New England Statistical Society},
  issn = {2693-7166},
  doi = {10.51387/24-NEJSDS66},
}

This repository is released under the MIT License. See LICENSE for details.

For questions, feedback, or bug reports, please open an issue or contact:

Joel Eliason
[email protected]