Skip to main content


Figure 12 | BMC Systems Biology

Figure 12

From: Supervised maximum-likelihood weighting of composite protein networks for complex prediction

Figure 12

Human BRCA1-A complex: PPI network, composite network, SWC-weighted network, and likelihood network. While the PPI network is fully connected, there are large numbers of extraneous edges and neighboring proteins (chiefly because BRCA1 itself is connected to around 180 proteins). In the SWC network, BRCA1 is still connected to a large number of proteins, but most of them are not connected to the other proteins in the complex, so they are unlikely to be clustered together. Clustering this network produces both the cluster consisting of the four complex proteins (generated by CMC), as well as a larger cluster consisting of the four complex proteins plus five additional proteins BABAM1, BRE, BRCC3, BRCA2, and FAM175B. Recent papers indicate that the former three additional proteins have been included in the BRCA1-A complex. The likelihood network shows that the three additional members are completely connected in a clique with two of the original complex members FAM175A and UIMC1 via PPI edges with strong co-complexness, while the four original members themselves are less strongly connected, via two functional associations with high co-complexness and a few low co-complexness PPIs.

Back to article page