Skip to main content


Fig. 9 | Journal of Cheminformatics

Fig. 9

From: An exploration strategy improves the diversity of de novo ligands using deep reinforcement learning: a case for the adenosine A2A receptor

Fig. 9

Comparison of the properties of generated molecules by RL models with different ε, β and Gφ. The pre-trained model on the ZINC set (a) and the fine-tuned model on the A2AR set (b) were used as Gφ. These properties included the number of hydrogen bond donors/acceptors, rotatable bonds, aliphatic rings, aromatic rings, and heterocycles

Back to article page