Table 2 Properties of gene interaction networks and modules derived from the networks under different similarity measures. All values are for the one-square matrix transformation method. See Methods and Discussion for detailed discussion, Figs. 3 and 4 for visual representation of the data, and supplementary online materials for generally similar results obtained under the two-square transformation

Similarity threshold applied to retain ~ 20,000 edges in the network

0.16

0.18

0.20

0.15

Nodes (genes) in the network / nodes (genes) in the giant connected component

3427 / 3303

3610 / 3587

3385 / 3321

4038 / 3956

Edges in the network / edges in the giant connected component

20,020 / 19,943

20,065 / 20,052

20,067 / 20,032

20,016 / 19,967

Unique genes in modules / percentage of all genes in respective giant connected component

2072 / 62.7

725 / 20.2

1519 / 45.7

3072 / 77.6

Number of modules / unique genes per module

682 / 3.04

408 / 1.78

516 / 2.94

1446 / 2.12

Biological Homogeneity Index

0.12

0.27

0.16

0.33

Percentage of uncharacterized genes / p-value / clusters with uncharacterized genes

36 / 10⁻³⁸ / 38

17 / 10⁻¹¹ / 2

36 / 10^− 22 / 30

26 / 10^− 3 / 35

Annotated modules of Type 1 / Type 2

36 / 409

64 / 119

48 / 309

279 / 568