Skip to content

README.md

Latest commit

 

History

History
44 lines (28 loc) · 1.65 KB

README.md

File metadata and controls

44 lines (28 loc) · 1.65 KB

Predicting microsatellite instability from whole slide images using texture features

This project utilizes texture features extracted from hematoxylin and eosin (H&E) stained whole slide images (WSIs) in order to predict microsatellite instability. Contact: Bahar Tercan btercan@systemsbiology.org
Reference:https://www.biorxiv.org/content/10.1101/2025.03.30.646246v1.full.pdf

Install and usage

Using the already segmented and MSI labelled tiles from the Kather dataset that can be found at DOI: 10.5281/zenodo.2530834 or at https://zenodo.org/records/2530835, the following example can be implemented.

In order to use XGBoost, an NVIDIA A100 Tensor Core GPU machine with CUDA toolkit enabled was used through Google Colab.

  1. Clone repository.
git clone https://github.com/IlyaLab/TextureAnalysis.git 
cd TextureAnalysis
  1. Install dependencies.
    NOTE: xgboost changes greatly between versions, so using xgboost==2.1.4 is the only guarenteed version that works for this example.
conda evn create -f environment.yaml
  1. In a Jupyter notebook, set the paths to tile directories, seperated by labels (already done in Kather dataset).
MSI_dir = r"/path/to/TCGA/CRC/MSI/tiles"
MSS_dir = r"/path/to/TCGA/CRC/MSS/tiles"
  1. Run main with the following parameters in order to predict MSI in TCGA-CRC (colorectal cancer) cohort. This will produce a per-tile ROC curve, a bar graph of the feature importance of the model, and a per-sample ROC curve.
main(MSI_dir, MSS_dir, sample_size=5000, n_jobs=-1, cohort_name='TCGA-CRC', validation=True, subsets=True, model='CRC', study='TCGA', ML='XGBoost')

License

Apache-2.0 license