Skip to main content

Table 3 Comparison of protein complex prediction by MIPS reference dataset

From: Finding low-conductance sets with dense interactions (FLCD) for better protein complex prediction

Network

Method

# complex

#. matched

Coverage

Recall

Precision

F-measure

Sn

PPV

Acc

MMR

SceDIP

FLCD

2134

120

3921

0.6522

0.1603

0.2573

0.4001

0.3901

0.3951

0.3206

 

CONE

380

74

1503

0.4022

0.1868

0.2551

0.2749

0.4015

0.3322

0.1533

 

LinkC

1839

109

3735

0.5924

0.1104

0.1862

0.4775

0.3646

0.4173

0.2993

 

SR-MCL

2851

41

4687

0.1964

0.0230

0.0402

0.4592

0.2104

0.3108

0.0726

SceBG

FLCD

4027

124

5836

0.6561

0.1393

0.2298

0.4643

0.4315

0.4476

0.3611

 

CONE

522

86

2735

0.4450

0.1533

0.2293

0.4537

0.4452

0.4494

0.1795

 

LinkC

5382

109

6076

0.6349

0.0918

0.1604

0.8179

0.3504

0.5354

0.3285

 

SR-MCL

1862

65

5889

0.3439

0.0673

0.1126

0.7360

0.2436

0.4234

0.1384

SceIntAct

FLCD

3394

120

4678

0.6417

0.1452

0.2368

0.4183

0.4034

0.4108

0.3482

 

CONE

496

79

1994

0.4225

0.1633

0.2356

0.3587

0.4296

0.3925

0.1927

 

LinkC

1297

80

5290

0.4278

0.0732

0.1251

0.9028

0.1986

0.4234

0.1886

 

SR-MCL

1079

45

5342

0.1337

0.0190

0.0941

0.6246

0.1850

0.3399

0.0960

SceMINT

FLCD

2483

111

4210

0.5904

0.1800

0.2759

0.4147

0.4086

0.4116

0.3231

 

CONE

513

67

2335

0.3564

0.1267

0.1869

0.3274

0.4017

0.3626

0.1519

 

LinkC

2201

100

4068

0.5319

0.1040

0.1740

0.4744

0.4038

0.4377

0.2744

 

SR-MCL

3698

24

4976

0.1277

0.0112

0.0205

0.4192

0.1999

0.2894

0.0481

  1. CONE and LinkC are short for ClusterONE and LinkComm, respectively
  2. Bold values denote the best scores corresponding to specific criteria