Fig. 6
From: Memory-assisted reinforcement learning for diverse molecular de novo design
ECFP6 analog generation during reinforcement learning with different bucket sizes. The different bucket sizes do not apply for the RL without a memory unit. In all figures the DRD2 QSAR model was used. a shows 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 shows the number of unique BM scaffolds of the generated ECFP6 analogs. c shows the number of unique carbon skeletons of the generated ECFP6 analogs