Skip to main content

Advertisement

Table 1 Seven QSAR datasets

From: Cross-validation pitfalls when selecting and assessing regression and classification models

Dataset Output Number of compounds Number of descriptors after preprocessing
AquaticTox Numeric 322 184
bbb2 2 Categories 79 22
Caco-PipelinePilotFP 3 Categories 3796 379
Caco-QuickProp 3 Categories 3796 47
MeltingPoint Numeric 4126 169
Mutagen 2 Categories 4335 1283
PLD 2 Categories 324 308
  1. Summary of 7 QSAR datasets.