Skip to main content

Table 6 Bayes information criterion (BIC) measure of Gaussian mixture model discretization used to fit gene-expression data for different \({\mathsf {k}}\) number of components

From: Boolean factor graph model for biological systems: the yeast cell-cycle network

Genes

BIC Scores (\(1\times 10^{3}\))

\({\mathsf {k}}=1\)

\({\mathsf {k}}=2\)

\({\mathsf {k}}=3\)

\({\mathsf {k}}=4\)

\({\mathsf {k}}=5\)

Cln3

2.6183

2.6020

2.6193

2.6397

2.6600

MBF

2.0492

1.8838

1.8913

1.9039

1.9182

SBF

2.0083

1.9608

1.9640

1.9781

1.9901

Cln1,2

2.6117

2.3853

2.3923

2.4117

2.4318

Cdh1

1.7429

1.4212

1.4066

1.4234

1.4404

Swi5

2.3690

2.1339

2.1354

2.1468

2.1660

Cdc20,14

1.7202

1.6175

1.6211

1.6326

1.6525

Clb5,6

2.0188

1.9799

1.9333

1.9374

1.9499

Sic1

1.4909

1.4785

1.4824

1.5027

1.5232

Clb1,2

2.5741

2.5267

2.5218

2.5330

2.5470

Mcm1,SFF

1.5210

1.4607

1.4757

1.4957

1.5123

  1. The lowest BIC value is the best fitting model, as highlighted in bold