Skip to main content

Table 2 Average ROC/ROC-50 scores on SCOP 1.67 benchmarks

From: Motif kernel generated by genetic programming improves remote homology and fold detection

Average ROC and ROC-50 scores

 

Superfamily level

Fold level

 

ROC

ROC-50

ROC

ROC-50

GPkernel

0.902

0.591

0.844

0.514

GPextended

0.869

0.542

0.753

0.371

GPboost

0.797

0.375

0.688

0.298

SVM-Pairwise

0.849

0.555

0.724

0.359

Mismatch

0.878

0.543

0.814

0.467

eMOTIF

0.857

0.551

0.698

0.308

LA-kernel

0.919

0.686

0.834

0.504

PSI-BLAST

0.575

0.175

0.501

0.010

  1. The table shows the average ROC and ROC-50 scores obtained by the different methods on the superfamily benchmark and the fold benchmark.
Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us