Fig. 7
From: Multi-objective de novo drug design with conditional graph generative model
Workflow for scaffold based molecule generation. Scaffold set S is first extracted from compounds in DrugBank. The conditional code \({\mathbf{c}}\) is set to be the substructure fingerprint based on S. Training is performed with the training samples labeled with \({\mathbf{c}}_G\). After training, scaffold based generation is performed using the fingerprint \({\mathbf{c}}_s\) of the query scaffold \(s \in S\)