Quantitative assessments. The bar graphs compare experimentally measured (a) growth rate, (b) H2 production, (c) PHB production and (d) CO2 production with predictions made by iRsp1095. Growth conditions included photoheterotrophic growth on the following carbon and nitrogen sources: succinate + glutamate (S-Gl), succinate + NH3 (S-N), glucose + glutamate (G-Gl), glutamate as a sole source of carbon and nitrogen (Gl only) and aerobic growth on succinate + ammonia (S-N (Aerobic)). The experimentally determined succinate uptake rate (SUR), ammonia uptake rate (AUR), glucose uptake rate (GUR) and glutamate uptake rate (GlUR) (all in mmol/g DW h) used in all simulations are provided in the boxes above the bars in Figure 3a. The FVA predicted maximum production rate of each of the metabolites during optimal growth (Max. predicted) is also provided. In each case the FVA predicted minimum production rate was zero. It should be noted that while ammonia can inhibit nitrogenase activity and thus H2 production, the ammonia uptake rates in S-N chemostats are not high enough to inhibit nitrogenase activity, hence the production of H2 under these conditions. No detectable amounts of PHB, H2 or CO2 were produced during growth on glutamate only, which is in line with model predictions. Figure 3e shows the decreases in predicted growth rate for S-Gl and S-N (Aerobic) conditions upon constraining the model with experimentally observed fluxes for H2, PHB and/or CO2 during simulations (Predicted (flux constrained)). The predicted optimal growth rates for other conditions shown in Figure 3a were unaffected by imposing these fluxes.