Skip to main content
Fig. 3 | Journal of Cheminformatics

Fig. 3

From: Memory-assisted reinforcement learning for diverse molecular de novo design

Fig. 3

ECFP6 analog generation during reinforcement learning. In a-c the HTR1A QSAR model was used. In d-f the DRD2 model was used. a and d show the number of generated ECFP6 analogs. Compounds with a prediction score of at least 0.7 and Tanimoto similarity (count-based ECFP6) to the nearest neighbor of known actives of at least 0.4 were considered ECFP6 analogs. b and e show the number of unique BM scaffolds of the generated ECFP6 analogs. c and f show the number of unique carbon skeletons of the generated ECFP6 analogs

Back to article page