An evolutionary computing approach for parameter estimation investigation of a model for cholera.

Area of Science:

• Computational Biology
• Epidemiology
• Mathematical Modeling

Background:

• Deterministic models are crucial for understanding disease dynamics but often require accurate parameter estimation.
• Time-series data provides valuable insights for refining these models.
• Parameter estimation in complex biological systems, like infectious diseases, presents significant challenges.

Purpose of the Study:

• To present a novel approach for parameter estimation in deterministic models using time-series data.
• To introduce genetic algorithms (GA) as an accessible tool for this purpose, requiring no specialized software.
• To demonstrate the application of GA for parameter estimation using a case study of cholera dynamics.

Main Methods:

• Development of a genetic algorithm (GA) implementation for parameter estimation.
• Utilizing time-series data from cholera outbreaks, specifically in Haiti.
• Applying the GA to estimate parameters for a deterministic cholera model.

Main Results:

• Successful implementation of a GA for parameter estimation without specialized toolboxes.
• Identification of multiple parameter sets that effectively describe the cholera time-series data.
• Demonstration of GA's capability to handle mechanistic uncertainty in disease models.

Conclusions:

• Genetic algorithms offer a flexible and accessible method for parameter estimation in deterministic models informed by time-series data.
• Comparing multiple parameter sets with similar performance can provide a more robust understanding of model behavior and underlying biological processes.
• This approach is particularly valuable for modeling infectious diseases with inherent mechanistic uncertainties, such as cholera.