Mathematical models are widely used to study biological processes by calibrating and comparing the models to experimental data. In this thesis, we provided methods for the robust and computationally efficient calibration of ordinary differential equation models. For this, we employed heavier tailed distribution assumptions for the measurement noise and exploited the structure of the underlying optimization problem. Moreover, we proposed a model for heterogeneous cell populations which is able to mechanistically describe and predict latent sources of cellular variability.
«
Mathematical models are widely used to study biological processes by calibrating and comparing the models to experimental data. In this thesis, we provided methods for the robust and computationally efficient calibration of ordinary differential equation models. For this, we employed heavier tailed distribution assumptions for the measurement noise and exploited the structure of the underlying optimization problem. Moreover, we proposed a model for heterogeneous cell populations which is able to...
»