Related Experiment Videos
Probabilistic methods for addressing uncertainty and variability in biological models: application to a toxicokinetic
1Center for Research in Scientific Computation, North Carolina State University, Raleigh, NC 27695-8205, USA. htbanks@eos.ncsu.edu
Mathematical Biosciences
|January 4, 2005
Summary
Biological systems exhibit variability and uncertainty, crucial for mathematical modeling. This study introduces probabilistic methods that better capture these population dynamics compared to deterministic approaches.
Area of Science:
- Mathematical biology
- Toxicology
- Computational modeling
Background:
- Biological systems inherently possess population variability and uncertainty.
- Accurate mathematical modeling requires accounting for these biological features.
- Standard deterministic methods may not fully represent population dynamics.
Purpose of the Study:
- To present probability-based parameter estimation methods for biological models.
- To address population variability and uncertainty in mathematical modeling.
- To evaluate the efficacy of probabilistic methods against deterministic ones.
Main Methods:
- Developed and discussed theoretical results for well-posedness and stability of probabilistic methods.
- Applied a probabilistic parameter estimation technique to a toxicokinetic model.
- Utilized simulated data for model application and comparison.
Main Results:
- Probabilistic methods demonstrated a superior ability to capture population variability.
- Uncertainty in model parameters was better represented using probabilistic approaches.
- The toxicokinetic model for trichloroethylene was analyzed using these methods.
Conclusions:
- Probabilistic parameter estimation is a robust approach for biological modeling.
- These methods offer significant advantages over deterministic techniques for systems with variability.
- The findings support the use of probabilistic methods in toxicology and systems biology.