Stoichiometric modeling represents a metabolic network as a matrix (the stoichiometric matrix S) where rows are metabolites, columns are reactions, and entries are stoichiometric coefficients. The steady-state mass balance condition — the rate of production equals consumption for each internal metabolite — is expressed as S * v = 0, where v is the vector of reaction fluxes. The solution space (null space of S) defines all thermodynamically and stoichiometrically feasible flux distributions. This framework is the mathematical foundation for constraint-based metabolic modeling and flux balance analysis.
Every metabolic reaction in a cell converts specific substrates into specific products in defined ratios — its stoichiometry. Glucose is split into two pyruvates, not three. Each turn of the TCA cycle consumes one acetyl-CoA and produces specific numbers of NADH, FADH2, and GTP molecules. Stoichiometric modeling takes these fixed ratios and assembles them into a comprehensive mathematical framework that describes the entire metabolic network simultaneously.
The central object is the stoichiometric matrix S. Each row represents a metabolite, each column represents a reaction, and each entry gives the stoichiometric coefficient — negative for substrates consumed, positive for products generated. For the reaction "A + 2B -> C", the column would have -1 in A's row, -2 in B's row, and +1 in C's row. The matrix captures the complete topology and mass-balance relationships of the network in a compact linear-algebraic form.
At metabolic steady state, no internal metabolite accumulates or depletes — the total rate of its production equals the total rate of its consumption. Mathematically, this is S * v = 0, where v is the vector of all reaction fluxes. This single matrix equation encodes the mass-balance constraint for every metabolite simultaneously. The set of all flux vectors satisfying this equation — the null space of S — defines the complete space of stoichiometrically feasible metabolic behaviors. Any flux distribution the cell could possibly adopt at steady state must lie within this space.
The power of this framework is its scalability. The stoichiometric matrix requires no kinetic parameters — only knowledge of which reactions exist and their stoichiometry, both of which are available from genome annotations and biochemical databases. This enabled construction of genome-scale metabolic models (GEMs) containing thousands of reactions for hundreds of organisms. The null space is typically high-dimensional (many feasible flux distributions exist), so additional constraints — reaction reversibility, measured uptake rates, thermodynamic feasibility, and optimization objectives — are layered on top of the stoichiometric framework. This constrained approach, formalized as flux balance analysis, has become the workhorse method of systems metabolic biology and metabolic engineering.