This repository contains the code and data for a machine learning approach to predict the binding affinity between a protein and a ligand.
Python 3.10 or higher
git
make 4.2.1
Clone this repository to your local machine:
git clone https://github.com/mahmodDAHOL/DTA.git
Navigate to the cloned repository:
cd DTA
Download the data from the here and extract it to the data directory, or you can run command
make install
this command will install all data and will create virtual enviroment with all required dependencies.
To verify that all thing will go well run:
make test
if all test passed, all things are ready.
To trainig the gnn run:
make run
this commad will train 5 folds, and create the following folders:
- logs: describe what are happening during run time
- models: contains 5 folders, one for each fold
- results: contains evaluation results on test data
- runs: tensorboard for visualizeing model architecture and metrics as graphs
After you got the models, you can run:
make clean
to remove data and virtual environment.
This project is licensed under the MIT License - see the LICENSE file for details.
If you have any questions or suggestions, please feel free to contact us at [email protected]. We welcome contributions and feedback from the community.