Skip to main content
. 2022 Jul 21;20:4040–4051. doi: 10.1016/j.csbj.2022.07.025

Fig. 6.

Fig. 6

Network approach uncovers TF complexes and hubs. a) The GM12878 co-occurrence network. Nodes represent TFs, edges represent associated TF pairs. Edges are colored by cosine score and sized by the number of associated TF1-TF2 co-occurring sites. Coloring of nodes illustrates Louvain community clustering. Two sub-clusters are indicated for zooming in c). b) GO-term analysis on TF groups extracted from network clusters (columns). Coloring indicates enrichment for GO terms (rows). c) Directional sub-networks from a). Each arrow represents a dependency defined by the confidence score, e.g. ATF3 is dependent on USF2. Node coloring indicates the number of sites assigned to the respective TF. Only edges with confidence scores above 0.4 are shown. d) Distribution of node degrees. High node degree is shaded with yellow in the plot. e) Number of connections per TF (blue), approximates power-law distribution (grey dashed line). f) Table of TFs per cell line with the highest number of interactions (hub creators) from d). Marked TFs (red) are discussed in the main text. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)