Transcriptomic profiling of shed cells enables spatial mapping of cellular turnover in human organs

Date 24/07/2025

The following figures are created using the following scripts:
Main figures
============
# Figure 2 - Turnover potential in the mouse small intestine.
>> fig2/01-shedding_mouse.ipynb
# Figure 3 - Turnover potential maps of the human upper GI tract.
>> fig3/A02-shedding_busslinger.ipynb
# Figure 4 - Turnover potential in the human colon. 
>> fig4/03-scatter_on_image.ipynb
>> fig4/04-knn_shedding_map.ipynb
# Figure 5 - Expression signatures and microenvionment of colonocytes with high turnover score.
>> fig4/01-analyze_olivera_short.ipynb
>> fig4/01-hgca_shedding.ipynb
>> fig4/03-scatter_on_image.ipynb

Expanded View figures
=====================
# Figure EV1 - Turnover scores of single cells from mouse small intestine.
>> ev_apoptosis/compare_apoptotic_genes.ipynb
# Figure EV2 - Reproducability and marker stratification of NGT fluids. 
>> fig3/A02-shedding_busslinger.ipynb
# Figure EV3 - Spatial turnoverv maps of the human small intestine. 
>> fig3/A03-analyze_harnik.ipynb
# Figure EV4 - Luminal shedding is different from basolateral shedding. 
>> ev_cfrna/compare_wash_and_blood.ipynb

Conda environment
=================
Setting an environment:
.yml file that was used on WSL is attached.
Use the following commands to set the environment
conda create -n turnover_env python=3.10
conda activate turnover_env
pip install "spatialdata[extra]"
pip install plotly adjusttext openpyxl igraph leidenalg