Table 5 AUC to predict binary target feasible/non feasible over the 30 selected molecules, coming from 3 similarity constrained generations
From: Integrating synthetic accessibility with AI-based generative drug design
Molecule name | RA score | SC score | SA score | RScore3min | RSPred |
|---|---|---|---|---|---|
Imatinib | 0.8 | 0.28 | 1 | 1 | 0.68 |
Acetylsalicylic acid | 0.44 | 0 | 0.33 | 1 | 0.222 |
Pi3K/mTOR | 0.76 | 0.4 | 0.96 | 1 | 0.48 |
Overall | 0.41 | 0.201 | 0.63 | 1 | 0.335 |