Fig. 6: Feature analysis of the NR1 CG compounds.
a Pie charts of the number of compounds per subfamily and the number of diverse skeletons per subfamily. b Potency distribution of the CG compounds is shown as the negative decadic logarithm of potency. Violin plots represent the potency distribution of all known ligands of the respective subfamily (≤ 100 µM from dataset20) and stars represent the selected CG compounds. c Pairwise Jaccard-Tanimoto similarity heatmap of the selected NR1 CG compounds computed on Morgan fingerprints. d t-SNE plot of known NR1 ligands (gray, ≤ 100 µM from dataset20) with the selected NR1 CG compounds highlighted and colored by their NR target subfamilies.
