STAD-CEACAM-scRNA

Figure generation code for: CEACAM5/6 as immunotherapy resistance markers in gastric cancer (single-cell RNA-seq study).

Environment Setup

Main environment (119/120 scripts)

conda env create -f environment.yml
conda activate stad_ceacam

Milo environment (Figure 2F only)

The Milo differential abundance analysis requires a separate environment with pertpy and R dependencies:

conda create -n pertpy_milo -c conda-forge python=3.11 -y
conda run -n pertpy_milo pip install pertpy scanpy matplotlib seaborn filelock
conda install -n pertpy_milo -c conda-forge -c bioconda rpy2 r-base bioconductor-edger bioconductor-limma r-statmod -y

BayesPrism (R, preparation scripts only)

conda create -n r_bayesprism -c conda-forge r-base r-data.table r-devtools -y
conda run -n r_bayesprism R -e 'if (!require("BiocManager")) install.packages("BiocManager"); BiocManager::install(c("NMF","scran")); devtools::install_github("Danko-Lab/BayesPrism/BayesPrism")'

Usage

Run all figure panels and assemblies:

bash 03_Final_Panels/_run_all_panels.sh

Run a single figure:

bash 03_Final_Panels/_run_all_panels.sh 3      # Figure 3 only
bash 03_Final_Panels/_run_all_panels.sh supp   # supplementaries only

External Databases

The SCENIC pipeline requires cisTarget motif databases (~95 MB, not bundled). Download into 02_Preparation_for_Panels/SCENIC/database/:

# cisTarget motifs database (Aerts lab)
wget https://resources.aertslab.org/cistarget/motif2tf/motifs-v10nr_clust-nr.hgnc-m0.001-o0.0.tbl \
  -O 02_Preparation_for_Panels/SCENIC/database/motifs-v10nr_clust-nr.hgnc-m0.001-o0.0.tbl

Data Availability

Input data (h5ad files, spatial data, external datasets) are available upon request or from the repositories described in the manuscript.

Citation

Manuscript in preparation.

License

MIT