QSAR Bioactivity Predictor is a Python application that allows users to create QSAR models to predict bioactivity for a specific target.

To use QSAR-Bioactivity-Predictor, you need to have Python 3.x installed on your system. You also need to install the following Python packages:

• chembl-webresource-client==0.10.8
• numpy==1.24.2
• pandas==1.5.3
• PyQt5==5.15.9
• PyQt5-Qt5==5.15.2
• rdkit==2022.9.5
• scikit-learn==1.2.2
• seaborn==0.12.2

You can install these packages using pip by running the following command:

  $ pip install -r requirements.txt

To run the QSAR-Bioactivity-Predictor application, open a terminal window and navigate to the directory where the qsar-predictor.py file is located. Then, run the following command:

  $ python qsar-predictor.py

This will open the main window of the application, where you can select a target and create a QSAR model to predict bioactivity for that target.

The QSAR-Bioactivity-Predictor application provides the following features:

• Load data using Chembl webresource client containing molecular descriptors and bioactivity values for a specific target.

• Preprocess the data by removing missing values and normalizing the descriptors.
• Train a QSAR model using random forest regressor
• Plot the experimental versus predicted pIC50 values.
• Predict pIC50 values of input SMILES CSV.

• Save the model to a file for later use.

If you want to contribute to QSAR-Bioactivity-Predictor, feel free to fork the repository and submit a pull request with your changes.

• The QSAR Bioactivity Predictor application was developed as a final project for a Master's degree.
• The application uses the PaDEL-Descriptor to calculate molecular descriptors from chemical structures.
• The QSAR models were trained using the scikit-learn library.

If you find this project useful in your research, please consider citing our paper:

Amine, A.M.E., Fadila, A. Transformer neural network for protein-specific drug discovery and validation using QSAR. J Proteins Proteom (2023). https://doi.org/10.1007/s42485-023-00124-6

@article{AmineFadila2023,
  author    = {Atil Mohamed El Amine, Atil Fadila},
  title     = {Transformer neural network for protein-specific drug discovery and validation using QSAR},
  journal   = {Journal of Proteins and Proteomics},
  year      = {2023},
  doi       = {10.1007/s42485-023-00124-6}
}

QSAR-Bioactivity-Predictor is licensed under the MIT License. See the LICENSE file for more information.