Skip to main content

Table 6 Comparison of performance against the state-of-the-art methods for remote homology detection

From: Efficient use of unlabeled data for protein sequence classification: a comparative study

Ā 

PDB

Swiss-Prot

NR

Ā 

ROC

ROC50

ROC

ROC50

ROC

ROC50

triple(1,3), full seq.

.9475

.7582

.9245

.6908

.9419

.7327

triple(1,3), region

.9708

.8265

.9752

.8556

.9824

.8861

triple(1,3), region, clustering

.9716

.8246

.9732

.8605

.9861

.8944

mismatch(5,1), full seq.

.9389

.7203

.9253

.6685

.9423

.7233

mismatch(5,1), region

.9698

.8048

.9757

.8280

.9837

.8824

mismatch(5,1), region, clustering

.9705

.8038

.9773

.8414

.9874

.8885

profile(5,7.5)

.9511

.7205

.9709

.7914

.9734

.8151