Evaluating the estimability of within-host population dynamics models
Despite the impacts of within-host disease dynamics on disease outcomes in individual hosts and disease spread among-hosts, generic models of within-host population dynamics have received far less attention than their among-host counterparts. While a number of models have been proposed to explore theoretical eco-evolutionary dynamics, they have yet to be evaluated for estimability, raising…
Within-host disease dynamics play a significant role in determining individual host outcomes and disease spread among hosts. However, generic models of within-host population dynamics have received less attention compared to their among-host counterparts. Various models have been proposed to investigate theoretical eco-evolutionary dynamics, yet their estimability has not been thoroughly examined, leading to concerns about their reliability in providing accurate inferences from data.
This study aimed to evaluate the estimability of two generic within-host population dynamics models by examining three key aspects: parameter estimation, the consequences of mis-assigning the underlying mechanistic model on parameter estimation, and reproducing qualitative dynamics. In some instances, fitting a model with mismatched structure to time series data resulted in reasonable parameter estimates that were capable of reproducing observed system dynamics.
Interestingly, even when provided with the actual data-generating model, substantial behavioral uncertainty could arise from the parameter uncertainty. These findings emphasize the influence of structural, parametric, and behavioral uncertainty on inference and demonstrate the importance of enhancing system-specific knowledge to avoid the application of incorrect functional forms and the measurement of consequential parameters to enhance estimability.
Written by urgent.news from bioRxiv's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
