Table 9 The comparison between condtional generative models RL based models in the task of generating dual inhibitors against GSK-3\(\beta\) and JNK3
From: Multi-objective de novo drug design with conditional graph generative model
Model | % valid | \(R_{\mathbf{c}}\) | \(EOR_{\mathbf{c}}\) | Diversity (\(I_{\mathbf{c}}\)) |
|---|---|---|---|---|
REINVENT | 0.999 ± 0.001 | 0.996 ± 0.005 | 1245 | 0.3 ± 0.2 |
Naive RL | 0.987 ± 0.007 | 0.969 ± 0.022 | 1211 | 0.4 ± 0.1 |
Graph | 0.955 ± 0.003 | 0.61 ± 0.00 | 759 | 0.814 ± 0.002 |
SMILES | 0.944 ± 0.003 | 0.56 ± 0.01 | 698 | 0.821 ± 0.001 |