ScaffoldGVAE

ScaffoldGVAE: A Variational Autoencoder Based on Multi-View Graph Neural Networks for Scaffold Generation and Scaffold Hopping of Drug Molecules

Installation

You can use the environment.yml file to create a new conda environment with all the necessary dependencies for ScaffoldGVAE.

git clone git@github.com:ecust-hc/ScaffoldGVAE.git
cd ScaffoldGVAE
conda env create -f environment.yml
conda activate ScaffoldGVAE

ScaffoldGVAE includes three sub-modules:

Sca_extraction.py: The molecular scaffolds was extracted and the data set was constructed.
Train.py: Used for pre-training on big dataset.
fine_tuning.py: Used for fine-tuning the pre-trained neural network on the known bioactive compounds against specific protein targets.
sample.py: Given a reference molecule and its corresponding scaffold ,sampling new scaffolds and adding side-chain for scaffold hopping.

Example of running the command:

python Sca_extraction.py

python pre_train.py

python fine_tuning.py

python sample.py

Name		Name	Last commit message	Last commit date
Latest commit History 8 Commits
.idea		.idea
.ipynb_checkpoints		.ipynb_checkpoints
GraphDTA		GraphDTA
Metrics		Metrics
__pycache__		__pycache__
figure		figure
models		models
.gitignore		.gitignore
LICENSE		LICENSE
README.md		README.md
Sca_extraction.py		Sca_extraction.py
Train.py		Train.py
add_side_chain.py		add_side_chain.py
data_structs.py		data_structs.py
draw_mole.ipynb		draw_mole.ipynb
environment.yaml		environment.yaml
fine_tuning.py		fine_tuning.py
model.py		model.py
sample.py		sample.py
t-sne.py		t-sne.py
utils.py		utils.py