Skip to main content

Table 3 The PC-LogicNet performance is evaluated for different values of c = α + β, i.e. c = 500, 750, 1000 and 1250. The PC-LogicNet is applied to reconstruct the directed network and logic functions among the regulatory genes in the E. coli, by using 10 gene expression samples and 100 repeats of the whole simulation study. Reported values for the TP, FP, TN, FN are the total of the corresponding values over 100 repeats of the whole simulation study

From: LogicNet: probabilistic continuous logics in reconstructing gene regulatory networks

 

Graph

TP

FP

TN

FN

TPR

FPR

PPV

ACC

MCC

F-measure

c = 500

Undirected a

716

165

2835

784

0.48

0.06

0.81

0.79

0.50

0.60

Directed b

616

592

6908

884

0.41

0.08

0.51

0.84

0.36

0.45

Directed logical c

614

592

–

885

0.41

–

0.51

–

–

0.45

c = 750

Undirected

720

155

2845

780

0.48

0.05

0.82

0.79

0.51

0.61

Directed

620

590

6910

880

0.41

0.08

0.51

0.84

0.37

0.46

Directed logical

605

590

–

899

0.40

–

0.51

–

–

0.45

c = 1000

Undirected

724

157

2843

776

0.48

0.05

0.82

0.79

0.51

0.61

Directed

624

588

6912

876

0.42

0.08

0.51

0.84

0.37

0.46

Directed logical

625

588

–

868

0.42

–

0.52

–

–

0.46

c = 1250

Undirected

758

147

2853

742

0.51

0.05

0.84

0.80

0.54

0.63

Directed

658

571

6929

842

0.44

0.08

0.54

0.84

0.39

0.48

Directed logical

632

571

–

845

0.43

–

0.53

–

–

0.47

  1. a The edge direction is not taken into account in calculating the performance
  2. b The edge direction is taken into account in calculating the performance
  3. c The integrative performance of the LogicNet in reconstructing both the edge direction and logic function among regulatory genes is evaluated