Agreement between constraint-based modeling predictions and experimental observations. (a) Assessment of constraint based modeling predictions for slow growth. Precision and recall (blue and green) were calculated for each model in each of 20 conditions (glycerol:* and glucose:ammonium excluded; Additional file 6) and means are shown here. The fraction of conditions in which predicted model mutants overlap significantly with significant z-score effects is shown in purple. (b) Precision and recall scores (as in (a)) for each individual condition using the Yeast5 model and standard FBA. (c) Number of producible metabolites for iMM905 and Yeast5 metabolic models. For each model the total number of producible metabolites was counted based on simulation in glucose:ammonium (see Materials and methods). The procedure was repeated for a model in which reactions involving auxotrophic marker genes (HIS3, URA3, LEU2, and MET15) were disabled. The chart shows the proportion of metabolites that the auxotrophic model fails to produce in red.