Figure 7
From: Inferring topology from clustering coefficients in protein-protein interaction networks
Influence of spurious interactions on clustering. Spurious interactions can influence average clustering coefficients in two ways depending on the degree of clustering in the network. In randomly (e.g. EX in A) or less than randomly clustered networks clustering coefficients can be increased by attaching false positive interactions. In highly clustered networks (e.g. GH in B), clustering coefficients are – at least for reasonable error rates – decreased. Here, average degree values were fixed at 10 and bait coverage at 1.