Table 5 Best feature selection technique with the optimal feature count.

From: GRUATT-AVP: leveraging a novel attention-based gated recurrent unit to advance the accuracy of antiviral peptide prediction

Feature selection techniques

Optimal

subset

AUC

ACC

MCC

Sen

Spe

Pre

F1

AUPR

PCA

400

0.975

0.929

0.859

0.942

0.916

0.922

0.932

0.977

LASSO

600

0.961

0.896

0.793

0.889

0.903

0.907

0.898

0.966

ICA

400

0.554

0.549

0.102

0.501

0.601

0.571

0.534

0.547

SHAP

500

0.986

0.948

0.896

0.942

0.954

0.956

0.949

0.988

Back to article page