Skip to main content

Table 3 Result of comparison test

From: Improved method for protein complex detection using bottleneck proteins

PPI network dataset

Reference dataset

Algorithm

Optimal parameters

Number of protein complexes

Recall

Precision

F1 score

DIP

MIPS

Proposed

CC = 0.9, BC = 1%

269

0.5556

0.3086

0.3968

  

[16]

CC = 0.51, BC = 20%

76

0.3210

0.4605

0.3783

  

Link Cluster

Partition_density = 0.30

1,177

0.7037

0.1427

0.2373

  

MCL

Granularity = 2.00

614

0.5679

0.0739

0.1298

  

MCODE

Node_score = 0.10

83

0.2930

0.2530

0.2729

 

CYC

Proposed

CC = 0.6, BC = 1%

646

0.4877

0.4860

0.4869

  

[16]

CC = 0.38, BC = 20%

333

0.3898

0.4114

0.4003

  

Link Cluster

Partition_density = 0.29

1,179

0.5932

0.2858

0.3857

  

MCL

Granularity = 2.40

639

0.4746

0.1690

0.2493

  

MCODE

Node_score = 0.10

83

0.2119

0.5542

0.3065

Bio-GRID

MIPS

Proposed.

CC = 1.0, BC = 1%

127

0.3457

0.4724

0.3709

  

[16]

CC = 0.54, BC = 20%

69

0.2346

0.3623

0.2848

  

Link Cluster

Partition_density = 0.30

10,463

0.5926

0.0893

0.1552

  

MCL

Granularity = 3.60

216

0.2099

0.0556

0.0879

  

MCODE

Node_score = 0.10

120

0.086

0.0500

0.0633

 

CYC

Proposed

CC = 1.0, BC = 15%

506

0.3260

0.3814

0.3515

  

[16]

CC = 0.43, BC = 30%

324

0.2500

0.2160

0.2318

  

Link Cluster

Partition_density = 0.28

10,915

0.5297

0.2802

0.3697

  

MCL

Granularity = 3.00

225

0.1144

0.1111

0.1127

  

MCODE

Node_score = 0.10

120

0.0593

0.1167

0.0787

I2D

CORUM

Proposed

CC = 0.8, BC = 5%

2,508

0.4100

0.3545

0.3802

  

[16]

CC = 0.41, BC = 20%

1,132

0.2961

0.2491

0.2706

  

Link Cluster

Partition_density = 0.21

8,033

0.4576

0.1595

0.2378

  

MCL

Granularity = 1.60

750

0.0623

0.0587

0.0604

  

MCODE

Node_score = 0.10

251

0.0469

0.1076

0.0652