Bayesian phylogeography finds its roots.

Area of Science:

• Molecular epidemiology
• Evolutionary biology
• Computational biology

Background:

• Geographical distribution of viruses is key to understanding endemic and epidemic dynamics.
• Previous methods for inferring viral migration patterns lacked temporal insights.
• Probabilistic models of evolution offer new statistical approaches for phylogeography.

Purpose of the Study:

• To introduce a Bayesian framework for inferring, visualizing, and testing viral phylogeographic history.
• To reconstruct timed viral dispersal patterns while accounting for phylogenetic uncertainty.
• To provide a versatile tool for molecular epidemiology and biogeographical inference.

Main Methods:

• Implementation of character mapping in Bayesian software sampling time-scaled phylogenies.
• Extension of standard Markov model inference with stochastic search variable selection.
• Utilization of virtual globe software for spatial and temporal data visualization.

Main Results:

• Reconstruction of timed viral dispersal patterns with phylogenetic uncertainty.
• Identification of parsimonious descriptions of viral diffusion processes.
• Demonstration of the framework's utility in analyzing influenza A H5N1 and rabies virus spread.

Conclusions:

• The developed Bayesian phylogeographic framework is a valuable asset for molecular epidemiology.
• The method can be generalized to infer biogeography from genetic data across various organisms.
• This approach provides deeper insights into virus origins, spread, and endemic maintenance.