Reproducible Research Archive

Haoran Chen, Yangyuan Zhang and Robert F. Murphy
Ray and Stephanie Lane Computational Biology Department
School of Computer Science
Carnegie Mellon University

June 16, 2025

This is the reproducible research archive that reproduces all results, figures and tables in the manuscript "CytoSpatio: Learning cell type spatial relationships using multirange, multitype point process models".

Intermediate results such as cell type assigning were also uploaded.

Run bash run_RRA.sh path_to_output to reproduce all results.

The "plotting" folder contains scripts to generate all figures and tables. The figure generation python scripts require seaborn==0.13.2 and pandas==2.3.0

Run bash makeallfigsandtables to generate PNGs. Some of these need to be assembly using Powerpoint to create the final figures in the manuscript.

Note that the script to create Figure 6 uses a saved data frame. If you want to recreate this dataframe from the results folders generated by CytoSpatio, download https://drive.google.com/drive/folders/1ez9ggyu6dA80QGoQ3MfJxiDP5uHX076S and modify plottingFig6Cto6FandS1.py to remove the comment in front of the call to builddataframe and insert comment in front of full_df = (see below)

build dataframe from the individual runs of Cytospatio

#builddataframe()

or the datasets are too large to download, you can use the checkpoint file to load the data

full_df = pd.read_csv('full_df.csv', index_col=0)