Modelling the dynamics of an experimental host-pathogen microcosm within a hierarchical Bayesian framework.

Area of Science:

• Ecology
• Epidemiology
• Biostatistics

Background:

• Bayesian statistical methods are widely used for analyzing infectious disease spread.
• Existing methods often fix parameter values, limiting the incorporation of prior information and uncertainty.
• Experimental host-pathogen systems present complex dynamics, multi-factorial designs, and measurement challenges.

Purpose of the Study:

• To introduce a Bayesian approach tailored for experimental host-pathogen systems.
• To leverage the flexibility of Bayesian methods for incorporating prior knowledge and parameter uncertainty.
• To analyze the dynamics of Paramecium caudatum and Holospora undulata infection.

Main Methods:

• Development and application of a hierarchical Bayesian model.
• Incorporation of non-linear dynamics, multi-factorial design, and sampling error.
• Utilizing prior distributions to account for existing information.

Main Results:

• Evidence for a saturable infection function in the host-pathogen system.
• Successful reproduction of two infection waves by distinguishing initial inoculum from subsequent releases.
• Inference of variations in the parasite's basic reproductive ratio across experimental groups.

Conclusions:

• The Bayesian framework effectively models complex host-pathogen dynamics.
• The approach allows predictions on how resources and host genotype affect parasite spread.
• This flexible Bayesian method has broad potential for applications in experimental ecology.