Skip to main content

Table 4 Performance on the testing dataset using the DPC molecular descriptor

From: MultiToxPred 1.0: a novel comprehensive tool for predicting 27 classes of protein toxins using an ensemble machine learning approach

Algorithm

ACC

F1

PPV

TPR

AUC

RF

0.779

0.777

0.773

0.783

0.97

MLP

0.794

0.791

0.787

0.799

0.98

XGBoost

0.784

0.784

0.782

0.788

0.99

LightGBM

0.787

0.788

0.787

0.791

0.99

LR

0.793

0.781

0.783

0.797

0.98

NB

0.667

0.645

0.682

0.670

0.95

k-NN

0.745

0.730

0.754

0.752

0.95

QDA

0.814

0.802

0.826

0.818

0.93