Such data sets often exhibit high variance with a paucity of replicates, thus providing a challenge for data mining. We describe data preprocessing and modeling methods that have proved reliable ...