Skip to main content
. 2012 Sep 6;8(9):e1002663. doi: 10.1371/journal.pcbi.1002663

Table 4. Edges present in the best-fitting models in human (BIC scores).

Number of edges Number of unique best-fitting models (Inline graphic) Basic set of edges Additional edges
1 4 (A,B)
2 5 (A,B), (B,D)
3 13 (A,B), (B,D), (A,C)
4 20 (A,B), (B,D), (A,C), (B,E) (E,C), (A,E), (A,D)
5 22 (A,B), (B,D), (A,C), (B,E), (E,C) (A,D)
6 25 (A,B), (B,D), (A,C), (B,E), (E,C), (A,D)
7 20 (A,B), (B,D), (A,C), (B,E), (E,C), (A,D), (A,E)
8 26 (A,B), (B,D), (A,C), (B,E), (E,C), (A,D), (A,E), (B,C)
9 15 (A,B), (B,D), (A,C), (B,E), (E,C), (A,D), (A,E), (B,C), (C,D)
10 1 (A,B), (B,D), (A,C), (B,E), (E,C), (A,D), (A,E), (B,C), (C,D), (E,D)

For each fixed number of edges Inline graphic , we report the basic set of edges (the most supported Inline graphic edges), as well as additional edges that are not significantly less supported (at Bonferroni-corrected significance level 0.05).