Skip to main content

Table 3 Function comparison: native recognition

From: LoCo: a novel main chain scoring function for protein structure prediction based on local coordinates

 

Ranknat

RMSDbest

Znat

CCnat

FEnat (%)

LoCo

13.4

1.62

1.805

0.519

36.6

DFMAC

6.7

1.17

2.630

0.562

38.3

RF_CB_SRS_OD

19.3

2.68

1.508

0.464

31.3

ProSa 2003

44.0

2.39

1.288

0.491

33.8

Four-body

81.8

4.87

0.621

0.334

20.4

General-four-body

56.3

4.67

0.797

0.311

18.9

Short-range

87.5

4.87

0.353

0.257

13.0

BFKV

54.5

3.54

0.774

0.397

24.7

BT

45.8

3.85

0.744

0.390

23.2

GKS

28.5

5.42

0.229

0.235

12.3

HLPL

31.4

3.37

0.602

0.383

24.8

MJ1

124.5

3.79

0.014

0.336

20.4

MJ2h

101.3

3.20

0.324

0.377

23.0

MJ3

50.6

4.69

0.401

0.244

15.7

MJ3h

52.1

3.63

0.733

0.410

26.3

MJPL

57.8

3.33

0.246

0.353

23.0

MS

54.0

4.94

0.419

0.234

13.3

MSBM

54.2

5.77

0.119

0.159

7.5

Qa

37.4

4.72

0.749

0.296

20.4

Qm

31.6

4.35

0.723

0.275

19.2

Qp

28.8

3.12

0.513

0.365

24.3

RO

248.3

5.67

0.287

0.248

17.6

SKJG

34.1

4.16

0.756

0.369

21.4

SKOa

33.1

4.35

0.664

0.352

20.5

SKOb

30.3

4.11

0.652

0.363

21.8

TD

47.7

3.81

0.739

0.399

24.2

Tel

80.0

4.03

0.740

0.370

23.5

TEs

54.2

4.50

0.646

0.331

17.2

TS

66.1

3.13

0.234

0.355

24.3

VD

73.7

5.09

0.478

0.290

17.5

  1. Native structure recognition performance comparison among scoring functions. All reported measures are averages over the 77 decoy sets in the final testing group. Lower scores are better for Ranknat and RMSDbest. Higher ones are better for Znat, CCnat and FEnat. LoCo outperforms all other functions except DFMAC in every measure. All metrics are defined in Performance measures at the end of Methods.