Fig. 8From: Multi-objective de novo drug design with conditional graph generative modelA visualized demonstration of model outputs. a–c. Output samples by MolRNN. Results are grouped by molecular weight (a MW < 300, b 300 \(\le\) MW < 500, c MW \(\ge\) 500); d, e Common mistakes made by SMILES based model and graph based model respectively; f Examples of broken aromaticity occurred during graph generationBack to article page