Skip to main content

Table 1 Logistic regression to predict the similarity, or otherwise, of training-set molecule-pairs using different types of fingerprint

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

Fingerprint β0 β1 R 2 t LR AUC
BCI −12.758 2.128 0.906 0.599 0.990
Daylight −10.677 1.850 0.884 0.577 0.986
ECFC4 −9.207 2.438 0.878 0.378 0.983
ECFP4 −12.754 2.524 0.894 0.505 0.988
MDL −9.022 1.380 0.812 0.654 0.973
Unity −12.347 1.956 0.884 0.631 0.987
  1. The columns contain the β0 and β1 values for the logistic regression model, the Nagelkerke R2 value, the computed value for t LR and the AUC for the ROC curve.