GenSSI

GenSSI is a MATLAB-based toolbox for structural identifiability analysis of dynamic models, particularly ordinary differential equation models used in systems biology, pharmacokinetics, and related fields. Structural identifiability questions whether model parameters can be uniquely inferred from perfect input-output data, independent of measurement noise or data quantity. GenSSI provides symbolic computation methods to determine identifiability by deriving the model’s input-output equations and examining algebraic relationships between parameters and observable quantities.

The toolbox is designed to handle models described by multiple state variables, parameters, and outputs, and it can accommodate various experimental setups such as different inputs or initial conditions. Users supply the model in terms of differential equations, state and output definitions, and the parameters to be analyzed. GenSSI then computes and reports the identifiability status for each parameter, classifying them as globally identifiable, locally identifiable (but not globally), or unidentifiable. It can also identify combinations of parameters that are structurally identifiable and suggest reparameterizations or additional experiments to improve identifiability.

GenSSI emphasizes a differential-algebraic approach, producing a structured analysis of identifiability that guides model reformulation and experimental design. It is widely used in research on model calibration, parameter estimation, and experimental planning. The software is released as a MATLAB package and has been distributed through sources such as GitHub and MATLAB File Exchange, reflecting its role as a practical tool for researchers working with complex dynamical models.