README.md

Codes for reproducing results in ``Bi-level Graph Learning Unveils Prognosis-Relevant Tumor Microenvironment Patterns from Breast Multiplexed Digital Pathology''

1. Install (If not done)

git clone https://github.com/JasmineZhen218/BiGraph4TME.git
cd BiGraph4TME
python3 -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt

2. Download Single Cell Data

• Discovery set

  1. Download cells.csv at https://drive.google.com/file/d/1TaMnyya2Lpa_s0CKPespBzEHXhUxU-ml/view?usp=drive_link
  2. Place it in Demo/Datasets/Danenberg_et_al

• External Validation set -1

  1. Download cells.csv at https://drive.google.com/file/d/1JpCFIVCNBWGSUVbmwjHQS1JgPOZhV1Ly/view?usp=drive_link
  2. Place it in Demo/Datasets/Jackson_et_al

• External Validation set -2 (optional):

  1. Get data access approval from https://zenodo.org/records/7990870
  2. Download NTPublic/data/derived/clinical.csv and NTPublic/data/derived/cells.csv
  3. Place them in Demo/Datasets/Wang_et_al.

3. Fit BiGraph Model

cd Demo
export OPENBLAS_NUM_THREADS=1 export GOTO_NUM_THREADS=1 export OMP_NUM_THREADS=1 python fit_all.py

4. Reproduce figures

python fig[X].py

Note: Run fig[X]_tnbc.py needs external validation set -2 exist in Demo/Datasets/Wang_et_al