Skip to main content

Advertisement

Table 2 Optimal levels of performance using ROC curves

From: The use of 2D fingerprint methods to support the assessment of structural similarity in orphan drug legislation

Fingerprint t ROC Probability Sensitivity Specificity Precision Accuracy F Youden Matthews
BCI 0.606 0.534 0.980 0.941 0.941 0.960 0.960 0.921 0.9208
Daylight 0.510 0.225 1.000 0.882 0.891 0.940 0.942 0.882 0.8866
ECFP4 0.490 0.406 0.980 0.922 0.923 0.950 0.951 0.901 0.9017
ECFC4 0.364 0.415 0.980 0.882 0.889 0.930 0.932 0.862 0.8645
MDL 0.650 0.487 0.939 0.882 0.885 0.910 0.911 0.821 0.8216
Unity 0.639 0.537 0.938 0.961 0.957 0.950 0.947 0.898 0.8990
  1. t ROC is the similarity threshold that gives the best level of performance, where this is that similarity value which maximises the values of the precision, the accuracy, the F index, the Youden index and the Matthews coefficient whilst maintaining acceptable values of the sensitivity and specificity. The largest values of these last five variables are bold-faced in the table.