Each differential equation described the rate of change in activity of a single species and the strength of its activity. Structurally, the models represented phospho-S400 and phospho-T525 as distinct species, and activation of Ste12 depended on their “concentration”. We simulated site mutants by setting the concentration of the phosphorylated species to 0 while keeping other parameter values constant. This approach facilitated simulation and exploration of different possible regulatory architectures in both reference and site mutant strains. We based the model on two assumptions. First, since in the Ste12 sequence both S400 and T525 precede proline residues, we assumed that the MAPK Fus3 phosphorylates these residues. Second, we assumed that phosphorylation of S400 and T525 and their effects on Ste12 were independent of each other. To explore the effects of the site mutants, we simulated both “wild type” and “mutant” models across the broad range of pheromone inputs for which we had experimental data. We selected generic parameter sets and analyzed the sensitivity of the model to each parameter. The simulated reference and mutant circuits recapitulated experimentally observed diminution of pathway output across a broad range of parameters. Thus, the data are consistent with a model in which phosphorylation of S400 and T525 increase the gain of the system. While many sites of phosphorylation have been mapped in proteomic studies from mammals, invertebrates and yeast, the vast majority of sites have no known function. We hypothesized that many of these sites are likely to exert dynamic regulatory roles in signaling pathways, the effects of which can only be revealed with quantitative assays in the context of the specific stimulus about which they convey information. We Paclitaxel therefore developed a systematic, general approach to prioritize the study of individual phosphorylation events and their potential functions in signaling networks, and demonstrated the utility of our approach on components of the pheromone response system in the budding yeast Saccharomyces cerevisiae. Since the approach described here requires targeted mutagenesis of genomic DNA, which is straightforward to accomplish in genetically PB 203580 tractable model organisms like yeast, we acknowledge that this methodology is more challenging to implement in other organisms. However with recent advances in tools like zinc finger nucleases and TALENs it will become increasingly possible to target mutations in many diverse organisms and cell types. We believe that the approach that we have employed here will soon be applicable to, e.g., iPS cells. We focused on three non-kinase proteins with distinct signaling roles upstream and downstream of the protein kinase cascade : 1) Ste12, a transcription factor activated by the MAP kinase cascade that induces genes involved in mating, 2) Ste50, an adaptor protein that acts upstream of the MAP kinase cascade to link the G protein-associated rho-like GTPase -PAK kinase complex to the MAP3K and, 3) Dig1, a protein involved in regulating the activity of Ste12. Using quantitative single cell assays we showed that 6 phosphorylation sites on 4 separate motifs in 3 signaling components quantitatively affected pheromone pathway output.