Skip to main content

Table 4 Performance comparison of MACCS and Morgan fingerprints in pharmacological class similarity search

From: Pharmacological affinity fingerprints derived from bioactivity data for the identification of designer drugs

 

MACCS

Morgan

EF10

AUC

Opt_thr

\({S}_{c}\)

EF10

AUC

Opt_thr

Stimulants

4.15 \(\pm\) 0.81

0.67 \(\pm\) 0.07

0.72 \(\pm\) 0.03

0.26

7.70 \(\pm\) 0.78

0.95 \(\pm\) 0.03

0.34 \(\pm\) 0.02

Cannabinoids

7.61 \(\pm\) 0.81

0.95 \(\pm\) 0.03

0.76 \(\pm\) 0.02

0.30

8.61 \(\pm\) 0.98

0.97 \(\pm\) 0.07

0.39 \(\pm\) 0.05

Serotonergic psychedelics

7.45 \(\pm\) 0.86

0.91 \(\pm\) 0.06

0.78 \(\pm\) 0.03

0.34

8.96 \(\pm\) 0.70

0.99 \(\pm\) 0.03

0.39 \(\pm\) 0.05

D-opioids

8.23 \(\pm\) 1.00

0.98 \(\pm\) 0.02

0.77 \(\pm\) 0.03

0.43

9.26 \(\pm\) 0.76

0.99 \(\pm\) 0.01

0.44 \(\pm\) 0.09

D-benzodiazepines

7.55 \(\pm\) 2.13

0.94 \(\pm\) 0.06

0.80 \(\pm\) 0.06

0.52

10.1 \(\pm\) 1.61

0.99 \(\pm\) 0.01

0.50 \(\pm\) 0.06

Average

7.00

0.89

  

8.92

0.98

 
  1. The data shown is the average of 50 similarity searches for each pharmacological class
  2. Both the query and test sets are composed of 1:10 active to decoy ratio by random sampling
  3. Opt_thr is the optimal threshold defined by the maximal G-Mean = \(\sqrt{Sensitivity\times Specificity}\)