Possibilistic estimation of metabolic fluxes during a batch process accounting for extracellular dynamics

Constraint-based models use the available knowledge about the operating constraints (e.g., mass balances and thermodynamic laws) to define a space of feasible states for cell cultures. Predictions can then be obtained incorporating experimental measurements of metabolite concentrations to perform a metabolic flux analysis. Although these predictions are typically static, aimed to study cells at given state, several works accounting for extracellular dynamics can be found in literature. In this work we formulate these predictions of time-varying fluxes and metabolites as possibilistic constraint satisfaction problems. The benefit of the described approach is that richer estimates are obtained —not only point-wise ones—, while considering uncertainty and even in scenarios of data Scirccity. The method could also be the basis for on-line fault detection in industrial processes.