Skip to main content

Table 2 Results of the decoration for each of the non-dataset scaffolds

From: SMILES-based deep generative scaffold decorator for de-novo drug design

S Generated ChEMBL decoys DRD2 decoys
Total % act. % act. % diff EOR % act. % diff EOR
6 1864 78.3 49.5 28.9 1.6 66.6 11.7 1.2
7 15,724 45.4 1.0 44.4 44.2 10.6 34.8 4.3
8 2178 80.2 44.3 35.9 1.8 49.3 30.9 1.6
9 5362 85.4 3.1 82.3 27.9 7.0 78.4 12.2
10 1012 98.9 90.4 8.4 1.1 93.7 5.2 1.1
  1. Total number of molecules sampled (Total); percent of generated molecules that are predicted as active (\(p_{active} \ge 0.5\)) by the APM (% act); For both the decoys decorated with ChEMBL fragments and DRD2 fragments from the training set: Percent of predicted active decoys (\(p_{active} \ge 0.5\)) (% act); difference between the generated predicted active percent and the predicted active percent of the decoys (% Diff); Enrichment Over Random (\(percent_{active} /percent_{decoy}\)) (EOR)