README.md
April 28, 2025 ยท View on GitHub
Running Steps
-
Prepare Input Files Ensure you have prepared the target protein's PDB file (e.g.
example.pdb). This file will serve as the basis for the input data. -
Data Preprocessing Use
data_utils.pyto extract structural information, SASA (Solvent Accessible Surface Area), and charge information from the PDB file, and save it as a CSV file. -
Modify Configuration Before running the prediction script, ensure the following settings are modified: Model Path: EMOCPD/models/model_save/best_6071model.pth.tar Input Data Path: Specify the path to the preprocessed CSV file. Output Path: Define where the prediction results should be stored.
Open the
predict.pyfile and modify the following code segment accordingly: MODEL_PATH = "EMOCPD/models/model_save/best_model.pth.tar" INPUT_CSV = "path/to/your/input.csv"
OUTPUT_DIR = "path/to/your/output.csv" -
Run Prediction Run the
predict.pyscript to generate the final prediction results.python predict.py
Output The prediction results will be saved in the specified output directory, typically as a CSV file, containing the following information: Predicted probabilities of amino acids at each position of the protein.