Fig. 1From: ScaffoldGVAE: scaffold generation and hopping of drug molecules via a variational autoencoder based on multi-view graph neural networksModel architecture diagram and workflow. A Multi-view graph neural network-based encoder. B RNN-based decoder for scaffold reconstruction and sampling. C Scaffold hopping: new scaffold sampling combined with reference molecule's side chains. D Flowchart depicting the training, sampling, and scaffold hopping of the modelBack to article page