The Science Behind Fermvyne
Flux balance analysis, kLa estimation, and constraint-based fed-batch optimization. Written for fermentation scientists and bioprocess engineers who want to understand the method — not just the output.
What is FBA and why does it work for aerobic fed-batch systems?
Flux Balance Analysis (FBA) is a constraint-based modeling framework that predicts the metabolic state of a cell by finding the flux distribution through its metabolic network that satisfies steady-state mass balance constraints and an assumed objective function (typically maximization of biomass growth or product formation).
Unlike kinetic models, FBA does not require knowledge of every enzyme kinetic parameter (Vmax, Km). Instead, it operates from the stoichiometry of metabolic reactions — data that is well-characterized for common industrial organisms and does not change with scale. This makes FBA well-suited for scale-up prediction: the stoichiometry remains constant, but the physical constraints (kLa, mixing, substrate gradients) change with vessel size.
Fermvyne uses a modified FBA approach that incorporates observed batch phenotype data (exchange fluxes inferred from your measured OD600, substrate consumption, and product titers) to constrain the solution space — making the model predictive for your specific strain under your specific operating conditions, not just a generic genome-scale model.
kLa estimation across vessel geometries.
The volumetric oxygen transfer coefficient (kLa) is the single most important physical parameter for aerobic fermentation scale-up. It determines whether your cells can get enough oxygen at the density and growth rate you're targeting.
How Fermvyne estimates kLa
Fermvyne uses empirical kLa correlations based on vessel geometry (impeller type, D/T ratio, number of impellers), operating conditions (superficial gas velocity, agitation rate), and media physical properties (viscosity, surface tension). The correlation set is calibrated for the most common impeller configurations used in industrial scale-up: Rushton turbine, pitched-blade turbine, and marine propeller.
The estimated kLa is then coupled to the OUR (oxygen uptake rate) predicted by the metabolic flux model at each growth phase. The resulting DO% vs time profile shows whether oxygen supply can keep pace with biological demand throughout the fed-batch — and at what point the agitation cascade needs to engage.
Scale-specific pressure correction
In large vessels (>2,000L), the hydrostatic pressure at the base of the vessel meaningfully increases dissolved oxygen partial pressure — which can temporarily mask an oxygen limitation that appears resolved by DO sensor readings near the top of the vessel. Fermvyne applies depth-dependent pressure correction to its kLa estimates for vessels above 2,000L.
Feed strategy and overflow metabolism prevention.
Acetate accumulation in E. coli
When glucose feed rate exceeds the maximum oxidative glucose uptake rate (q_glc_max), E. coli diverts carbon through the overflow metabolism pathway, secreting acetate. Acetate is toxic above ~2 g/L, inhibits growth, and can cause a rapid titer collapse mid-fed-batch.
Fermvyne's flux model predicts the critical specific glucose uptake rate for your strain at each OD600 setpoint. The feed rate protocol it generates stays below this threshold throughout the fed-batch, using exponential feeding in the growth phase and DO-stat control as the culture reaches high cell density.
Fumaric acid accumulation in mycoprotein systems
Fusarium venenatum (the organism used in Quorn-style mycoprotein fermentation) accumulates fumaric and malic acids under oxygen-limited conditions. Unlike acetate in E. coli, this overflow is driven by TCA cycle imbalance under low DO rather than substrate overflow at the glycolytic entry.
Fermvyne models this pathway explicitly for mycoprotein systems, providing a DO maintenance setpoint below which fumarate accumulation becomes significant — a parameter that shifts substantially between pilot and commercial scale due to kLa reduction.
Methanol feed modeling in P. pastoris
Methanol-induction phase in Pichia pastoris requires precise control of methanol feed to maintain induction without triggering methanol toxicity or hypoxia. Fermvyne models the methanol oxidation pathway, calculating the oxygen demand per mole of methanol oxidized and flagging the DO depletion risk at the methanol feed rates needed to achieve target volumetric productivity.
Feeding strategy comparison
For each simulation job, Fermvyne evaluates three feeding strategies and returns a comparative titer and DO prediction: (1) exponential feed at target specific growth rate, (2) linear ramp, and (3) DO-stat feed with proportional control. You select the strategy that best fits your DCS control loop capabilities.
Scientific foundations.
Fermvyne is built on established FBA and bioprocess engineering methodology. Key references for the modeling approaches we use:
Varma A, Palsson BO. Stoichiometric flux balance models quantitatively predict growth and metabolic by-product secretion in wild-type Escherichia coli W3110. Applied and Environmental Microbiology. 1994;60(10):3724–3731.
Palsson BO. Systems Biology: Constraint-based Reconstruction and Analysis. Cambridge University Press; 2015.
Sonnleitner B, Käppeli O. Growth of Saccharomyces cerevisiae is controlled by its limited respiratory capacity: formulation and verification of a hypothesis. Biotechnology and Bioengineering. 1986;28(6):927–937.
Garcia-Ochoa F, Gomez E. Bioreactor scale-up and oxygen transfer rate in microbial processes: an overview. Biotechnology Advances. 2009;27(2):153–176.
Orth JD, Thiele I, Palsson BO. What is flux balance analysis? Nature Biotechnology. 2010;28(3):245–248.
Explore our application notes.
Detailed technical writeups on E. coli DO control, P. pastoris methanol feed optimization, and mycoprotein titer assurance — with worked examples from our internal validation dataset.