Fig. 2.

Comparison between PROM (Top) and RFBA (Bottom): a perturbation to a TF results in alteration in expression of its target genes. These are then mapped onto the metabolic network. Depending on the gene state, the fluxes through the reactions are constrained and the optimal growth rate is determined by using FBA. In PROM, the constraints based on gene expression (red) are used as cues to obtain the optimal flux state, whereas in RFBA, genes and fluxes can have only two states (on/off). Further, PROM's automated inference of interactions and probabilistic formalism enables it to create comprehensive models.