Skip to main content

Table 3 Performance comparison of SV caller on HG00514 and HG00733 data

From: BreakNet: detecting deletions using long reads and a deep learning approach

   

BreakNet

Sniffles

cuteSV

SVIM

HG00514

42X

Precision

0.7197

0.6772

0.4539

0.5547

  

Recall

0.351

0.3137

0.3286

0.2261

  

F1

0.4721

0.4290

0.3811

0.3213

 

21X

Precision

0.5900

0.7358

0.4660

0.4407

  

Recall

0.2624

0.2898

0.3442

0.1632

  

F1

0.3633

0.4158

0.3960

0.2382

 

10X

Precision

0.5552

0.5986

0.5562

0.2593

  

Recall

0.3310

0.2585

0.3013

0.3003

  

F1

0.4147

0.3611

0.3909

0.2783

HG00733

48X

Precision

0.7160

0.6528

0.4834

0.5166

  

Recall

0.3578

0.3076

0.2932

0.2277

  

F1

0.4772

0.4182

0.3650

0.3162

 

24X

Precision

0.7598

0.7197

0.4250

0.5981

  

Recall

0.3386

0.2832

0.3379

0.2218

  

F1

0.4685

0.4065

0.3765

0.3235

 

10X

Precision

0.5444

0.6440

0.5117

0.2361

  

Recall

0.3364

0.2510

0.2959

0.3147

  

F1

0.4158

0.3611

0.375

0.2700

  1. Bold values represent best results