Skip to main content
. 2021 May 5;13:37. doi: 10.1186/s13321-021-00514-2

Fig. 6.

Fig. 6

PCA visualization of SANCDB and analogs chemical space. Compounds (n = 375,061) are represented in dots. SANCDB (violet, n = 1012). Analogs are in bins of similarity values: (0.6,0.7) blue, n = 266,147; (0.7,0.8) orange, n = 69,336; (0.8,0.9) green, n = 24,679; [0.9,1] red, n = 13,887. As an analog may have different similarity scores with different SANCDB compounds, the maximum similarity score was chosen for each analog. The first two components explain 81% of the variance (PC1 (66%), PC1 (15%))