Skip to main content
. 2017 Mar 9;7:44040. doi: 10.1038/srep44040

Table 1. Assessment results for each pharmacophore model.

Model Hta Hab A (%)c Y (%)d Ne CAIf
Model_01 173 90 90.00 52.02 1.85 1.67
Model_02 161 90 90.00 55.90 1.99 1.79
Model_03 163 89 89.00 54.60 1.94 1.73
Model_04 166 91 91.00 54.82 1.95 1.78
Model_05 167 90 90.00 53.89 1.92 1.73
Model_06 169 91 91.00 53.85 1.92 1.74
Model_07 180 91 91.00 50.56 1.80 1.64
Model_08 172 91 91.00 52.91 1.88 1.71
Model_09 168 91 91.00 54.17 1.93 1.75
Model_10 155 92 92.00 59.35 2.11 1.94

aHt is the number of hits.

bHa is the number of active hits.

cA% represents the ability to identify active compounds from the external database (A% = Ha/A, while A is the number of active compounds in the external database).

dY% represents the proportion of active hits in total hits (Y% = Ha/Ht).

eN represents the ability to distinguish active compounds from non-active compounds (N = (Ha/Ht)/(A/D), while D is the number of compounds in the external database).

fCAI was utilized to evaluate the models comprehensively (CAI = N × A%).