Skip to main content

Table 7 Average ranks based on the disease-associations for the best performing networks of each inference method

From: Functional networks inference from rule-based machine learning models

Source

Category

FuNeL

Pearson

ARACNE

MIC

Curated

1A

C2 (1)

SN(C2) (4)

SN(C3) (2.5)

SN(C2) (2.5)

 

2A

C3 (1)

SN(C3) (2)

SE(C2) (3)

SN(C2) (4)

 

3A

C1 (2)

SN(C1) (3)

SE(C4) (4)

SE(C2) (1)

 

Proximity

C2 (1)

SN(C3) (2.5)

SE(C4) (2.5)

SE(C2) (4)

 

Average

1.25

2.88

3

2.88

Malacards

1A

C2 (1)

SN(C2) (4)

SN(C4) (3)

SE(C4) (2)

 

2A

C2 (1.5)

SN(C4) (4)

SE(C4) (1.5)

SN(C2) (3)

 

3A

C3 (2)

SN(C4) (3)

SE(C2) (4)

SN(C2) (1)

 

Proximity

C2 (1)

SE(C4) (4)

SE(C4) (3)

SE(C2) (2)

 

Average

1.78

3.75

2.88

2

  1. For each category and for each method we report the network used for the analysis. The ranks (in brackets) were averaged across all 8 datasets, and the highest ranks are shown with bold font. The Average row reports the average ranks across all the categories. The number of disease-associated genes participating in a triangle is denoted as 1A, 2A and 3A