A molecular generation model based on deep learning algorithm. The generation task is implemented in the TensorFlow framework, allowing the user to run the model to generate a focused library of drug-like molecules.

⚙ Requirement

Refer to requirement.txt

🔧 Installation

• Install python 3.7 in Linux or Windows.
• If you want to run on a GPU, you will need to install CUDA and cuDNN, please refer to their websites for corresponding versions.
• Add the installation path and run the following command to install all the environment libraries in one step.

pip install -r requirement.txt

✈ Running this Model

You need to open main.py, run load_weights to read the pre-trained weights and get the generated molecules. Or provide training set molecules into coding for model training.

💥 Dataset

Folder datasets contains pre-training or transfer learning molecules, folder generate contains the molecules generated after transfer learning, folder gen_data is based on the generation model or machine learning model screening molecules.

📖 Article

For more details about the methodology and experimental results, please refer to the paper:

Generation of focused drug molecule library using recurrent neural network
DOI: 10.1007/s00894-023-05772-5