The notebooks: final_HX1 en final_E_coli are used to process the data and train the models. The notebooks calibration_HX1 and calibration_Ecoli are used to calibrate the trained models. The final notebook: annotation_analyses is used to perform clustering analyses and to asses model performance in differently annotated regions.
The convolutional neural network can be found in ConvNeXt_model_110.py, the transformer can be found in Transformer_model_window16.py.