FBA assumes cells optimize biomass production. When would this assumption fail, and how would you know?
AIt never fails — all cells always maximize growth rate
BIt fails when cells are in non-growth states (quiescence, stress response, stationary phase), detectable when FBA predicts growth but the cells are not actually growing
CIt fails only for eukaryotic cells because they are too complex
DIt fails only when the stoichiometric matrix has errors
The biomass objective function assumes cells have been evolutionarily selected to maximize growth rate — reasonable for microbes in exponential growth but not for cells in stationary phase, under stress, or performing specialized functions (like immune cells or neurons). FBA with a biomass objective would predict nonzero growth for any condition with available nutrients, which is incorrect for quiescent cells. Alternative objectives (minimizing metabolic adjustment, maximizing ATP yield, or multi-objective formulations) have been developed for these cases. The discrepancy between predicted and observed growth rates signals that the objective function needs revision.
Question 2 True / False
FBA requires detailed kinetic parameters (Km, Vmax) for every enzyme in the metabolic network.
TTrue
FFalse
Answer: False
FBA explicitly avoids kinetic parameters — this is its key advantage and the reason it scales to genome-scale models. It uses only stoichiometric constraints (S * v = 0), reaction bounds (minimum and maximum flux through each reaction, reflecting reversibility and capacity), and an objective function. The trade-off is that FBA predicts optimal steady-state flux distributions but cannot predict metabolite concentrations, transient dynamics, or allosteric regulation. When kinetic information is available for a subset of reactions, it can be incorporated as tighter flux bounds.
Question 3 Short Answer
A gene knockout is predicted by FBA to be lethal (zero biomass flux), but the organism survives in the lab. Name two biological explanations for this discrepancy.
Think about your answer, then reveal below.
Model answer: First, the metabolic model may be incomplete — the organism may have an alternative pathway not included in the reconstruction that bypasses the deleted reaction. Second, the organism may have adapted (evolved) to reroute flux through suboptimal but viable alternative pathways that FBA's strict optimality assumption does not predict; methods like MOMA (Minimization of Metabolic Adjustment) better capture this sub-optimal post-knockout behavior. Additional possibilities include isozymes not annotated in the model or regulatory changes that activate latent pathways.
Genome-scale models are never complete, and FBA's optimality assumption means it explores only a fraction of the feasible flux space. Experimental evolution studies have shown that organisms frequently find metabolic workarounds that were not predicted by FBA, often involving low-activity promiscuous enzymes or regulatory rewiring.