Figure 4.

Screenshot of CytoCopteR, the Cytoscape plugin for CellNOptR. Users can load or build a network in Cytoscape and load a matching data set in the MIDAS format, i.e. a CSV file with a row for each condition/time combination, a “TR:” column for each stimuli/inhibitor (0=absent,1=present) and for each readout a “DA:” column (time) and a “DV:” column (measurement). CytoCopteR annotates the original network with an overlaid color code on the edges and nodes (see subfigure A, left) reflecting the experimental (stimulated, inhibited, measured) and pre-processing (compressed or not) status for the nodes. Users then train the model to data, currently using the Boolean steady-state implementation in CellNOptR. The parameters for the training can be changed through explicit panels such as the one on subfigure B. Results of the pipeline are reported as in CellNOptR, via a graph displaying experimental and simulated data overlaid (see panel C), plots of the evolution of fit during the training process and diverse information of the training process (not shown). Furthermore, the scaffold network (after compression and expansion of the original network) is represented as a cytoscape network, with the same overlaid color code (see panel subfigure A, right) and weighting the edges according to their presence in the family of models retrieved.

Terfve et al. BMC Systems Biology 2012 6:133   doi:10.1186/1752-0509-6-133
Download authors' original image