Skip to main content

Table 2 Performance of different versions of ScanBMA on the yeast data, either on the original scale (Orig) or transformed (Trans)

From: Fast Bayesian inference for gene regulatory networks using ScanBMA

Data

nvar

Priors

Precision

AUROC

AUPRC

TP

FP

Trans

20

g, inform

0.391

0.601

0.0747

227

353

Trans

3556

g, inform

0.274

0.629

0.0740

127

336

Trans

20

BIC, inform

0.244

0.590

0.0616

200

619

Orig

20

g, inform

0.274

0.586

0.0680

552

1460

Trans

20

g, Guelzim

0.175

0.499

0.0395

34

160

  1. For the priors, inform refers to the informative prior, while Guelzim refers to the prior probability of 2.76/6000 for all possible relationships.
Back to article page

BMC Systems Biology

ISSN: 1752-0509

Contact us