Skip to main content
Fig. 2 | BMC Bioinformatics

Fig. 2

From: CABeRNET: a Cytoscape app for augmented Boolean models of gene regulatory NETworks

Fig. 2

Dynamical simulation and robustness analysis of an augmented T-helper GRN with CABERNET. a The T-helper signaling network, mapped in [20]. Edges stand for regulatory interactions, either activating (black) or inhibiting (red). The network is composed by 40 genes and 51 interactions. b The augmented NRNB that displayed a differentiation tree matching the hematopoietic one. To find it, 600 NRBNs were randomly generated by augmenting the T-helper GRN in CABERNET; the augmented networks include 200 nodes (160 nodes added to the original core) and 400 edges (349 new ones, average connectivity = 2). The nodes are wired according to a random Erdos-Renyi topology, and random Boolean functions with bias = 0.5 are associated to the nodes. Only matching NRBN is shown, the original core and the augmented portion of which are highlighted. In CABERNET’s visualization the size of each node is proportional to its connectivity degree and the color-scale to the function bias. c The Attractor Transition Matrix of the matching NRBN is plot by CABERNET, highlighting the noise-induced transitions among attractors and the Threshold Ergodic Sets representing cell types. The progressive splitting of the TESs due to increasingly larger noise resistance-related thresholds (i.e., δ=0,0.023,0.056,1) is shown, stressing the perfect matching between the emergent differentiation tree and that of hematopoietic cells, from multi-potent cells to fully differentiated cell types. d The differentiation tree of hematopoietic cells from [60] is depicted. Notice that T-helper cell type represents one of the leaves of the tree. For the description of the acronyms please refer to the main text. e Configuration of the 8 attractors of the augmented network (determining the gene activation patterns). In this specific case, the length of each attractor is equal to 8. f Robustness analysis performed via CABERNET. Single node knockout experiments (i.e., silencing the node’s Boolean function) are performed on each node of the original core of the augmented network and the dynamics is simulated again via CABERNET. The emergent tree is then compared with that of hematopoietic cells and the distribution of the similarity measure \(\widehat {d}\) (Eq. 1) is displayed, highlighting 5 genes that, when silenced, still lead to a matching emergent tree (i.e., \(\widehat {d}= 0\))

Back to article page