Dr. Clemens Kreutz
Uncertainty analyses in the systems biology using the profile likelihood
Was |
|
---|---|
Wann |
10.01.2014 von 11:30 bis 13:00 |
Wo | Eckerstr.1, Room 404 |
Termin übernehmen |
vCal iCal |
Understanding of biological processes by means of mathematical models is a major aim in systems biology. As an example, biochemical reaction networks are translated to ordinary differential equation models which are discriminated and calibrated based on experimental data. Since the models are nonlinear and contain a large number of parameters, classical statistical approaches for the calculation of confidence regions for parameter estimates and for model predictions are hardly feasible.
In this talk, the capacity of profile-likelihood based methodologies for uncertainty analyses is discussed. The methods are utilized to generate reliable confidence intervals for parameter estimates and model predictions, as well as for data-based identifiability and observability analyses. Moreover, it can be applied for experimental design optimization.
The presented approach constitutes a general concept and is therefore applicable for any kind of prediction model.