Table 2 Evaluation of TEMPROT and TEMPROT+ compared to the state-of-the-art approaches on the test set using \(F_{\max }\), AuPRC, IAuPRC and \(S_{\min }\) evaluation metrics

Naive

0.402

0.611

0.446

0.266

0.521

0.228

0.345

0.634

0.370

25.423

10.268

9.349

CAFA-BLASTp

0.468

0.469

0.551

0.208

0.215

0.287

0.215

0.216

0.296

38.083

18.755

9.124

DeepGO

0.337

0.379

0.489

0.247

0.257

0.309

0.304

0.382

0.465

27.414

11.880

8.821

DeepGOPlusCNN

0.498

0.664

0.531

0.444

0.637

0.460

0.465

0.634

0.528

23.799

9.783

8.240

TALE+Transformers

0.491

0.661

0.550

0.477

0.613

0.444

0.469

0.706

0.549

23.929

9.682

8.115

ATGO

0.547

0.684

0.616

0.506

0.667

0.623

0.524

0.724

0.632

22.228

9.437

7.228

TEMPROT

0.499

0.689

0.643

0.459

0.639

0.561

0.483

0.719

0.664

23.652

9.209

6.973

DIAMOND

0.519

0.593

0.572

0.286

0.237

0.320

0.417

0.483

0.462

23.066

9.957

7.164

BLASTp

0.561

0.637

0.620

0.402

0.380

0.360

0.502

0.586

0.562

22.183

9.795

6.805

DeepGOPlus

0.553

0.677

0.619

0.514

0.638

0.559

0.536

0.717

0.635

22.648

9.515

7.090

TALE+

0.555

0.681

0.631

0.547

0.643

0.621

0.540

0.724

0.643

22.615

9.363

6.949

ATGO+

0.589

0.690

0.652

0.550

0.660

0.650

0.571

0.731

0.689

21.233

9.286

6.617

TEMPROT+

0.581

0.692

0.662

0.529

0.641

0.595

0.558

0.728

0.689

21.892

9.169

6.662