Using Optimal Transformations and Multi-Experiment Fitting to Detect and Reduce Effects of Non-Identifiable Parameters in Non-Linear Dynamical Models

Abstract

Mathematical models of the dynamics of cellular processes promise to yield new insights into the underlying cell biology and their systems' properties. Since the processes are usually high-dimensional and time-resolved experimental data of the processes are sparse, parameter estimation faces the challenges of structural and practical non-identifiability of the parameters. Non-identifiabilities might render the systems analysis of the model difficult. Non-identifiability results usually in non-linear dependencies of the estimated parameters.