lung

lung

First download data from TCGA using

mkdir -p data && mv manifest.txt data/. && cd data
gdc-client download -m manifest.txt

Run lung.ipynb to preprocess data

Go trough predictor_LUAD_LUSC.ipynb to execute the predictor discussed in the paper, this can be run on Colab or in any machine with tensorflow (possibly with GPU enabled)

model_cancer.type.h5 contains the trained weights. To reproduce our evaluation analyses load it using

from tensorflow.keras.models import load_model
model = load_model(f"model_cancer.type.h5")
model.summary()

as described in the notebook.

Output for hierarchical stochastic block model is in the topsbm folder