Skip to main content

Table 2 Comparison of the top 3 property scores of generated molecules found by each model

From: DeepGraphMolGen, a multi-objective, computational strategy for generating molecules with desirable properties: a graph convolution and reinforcement learning approach

Method Penalized logP QED
1st 2nd 3rd Validity 1st 2nd 3rd Validity
ZINC 4.52 4.30 4.23 100% 0.948 0.948 0.948 100%
ORGAN 3.63 3.49 3.44 0.4% 0.896 0.824 0.820 2.2%
JT-VAE 5.30 4.93 4.49 100% 0.925 0.911 0.910 100%
GCPN 7.98 7.85 7.80 100% 0.948 0.947 0.946 100%
  1. Validity is defined as the fraction of generated molecules that are chemically valid. ORGAN and JT-VAE are described in [79] and [63], respectively
  2. Italics values refer to the best results among the methods compared