Skip to main content

Table 3 The first two rows show the average entropy for the networks and, for comparison, the average entropy for the networks with GO labels randomly permuted

From: Assessing the functional coherence of modules found in multiple-evidence networks from Arabidopsis

  ALL SEQ COE PPI LIT
Average entropy (actual network) 2.72 2.96 3.30 2.86 2.96
Average entropy (randomly permuted network) 3.31 3.43 3.42 3.29 3.33
Relative decrease in entropy (compared to randomly permuted network) 17.8% 13.7% 3.5% 13.1% 11.1%
BFRP 49.43% 22.78% 3.55% 24.56% 28.43%
BERP 39.58% 16.64% 2.75% 18.58% 31.18%
  1. The third row details the decrease in entropy observed in the actual network compared to the randomly permuted network. By both methods of measuring network fragmentation, the ALL dataset produces the best quality networks. The percentages may not add up to a 100%, as in cases where several networks performed the same all were counted as "best" for that GO term. The fourth and fifth rows show the best fragment rank percentage and best entropy rank percentage statistics (defined in the Methods section).