Table 3 Single-view features performance.

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

Features Extractors

Optimal strategy

AUC

ACC

MCC

Sen

Spe

Pre

F1

AUPR

ProtBert

Mean pooling

0.955

0.890

0.780

0.879

0.901

0.903

0.891

0.960

OHE

3 kmer

0.937

0.864

0.730

0.849

0.881

0.883

0.865

0.945

Glove

3 kmer

0.926

0.853

0.708

0.824

0.883

0.882

0.852

0.934

FastText

3 kmer

0.920

0.852

0.708

0.805

0.901

0.896

0.848

0.925

Word2vec

5 kmer

0.739

0.699

0.401

0.664

0.736

0.727

0.694

0.755

ESM

–

0.895

0.854

0.708

0.839

0.869

0.867

0.853

0.878

ProtT5

–

0.934

0.890

0.780

0.903

0.877

0.884

0.893

0.918

Back to article page