Skip to main content

Table 2 The percentages of generated networks with scores better than one of the original network. In the options of the algorithm with forbidden or forced self-loops, the original network was modified by removing or adding appropriate edges, respectively.

From: Applying dynamic Bayesian networks to perturbed gene expression data

dataset

all regulatory interactions

transcriptional regulation only

perturbations

time points

s-l forb.

s-l perm.

s-l forc.

s-l forb.

s-l perm.

s-l forc.

-

12

12.642

24.309

53.245

02.925

02.876

20.842

knockout

12

00.000

00.009

03.901

00.015

00.119

05.328

knockout

3

00.000

00.001

08.661

00.001

00.000

04.330

both

3

00.000

00.003

09.642

00.000

00.005

04.618