Table 4: The comparison of predictive performance between our method and GPS-PUP using the leave-one-out cross validation on Dataset 2.

Prediction methodSensitivity (%)Specificity (%)Accuracy (%)MCC

GPS-PUP63.7880.2178.850.2860.708
SinglePup73.9178.9576.190.5260.825
EnsemblePup80.0083.3382.000.6290.873