An integrative modelling approach that combines pathogen genetic sequence data (phylogenies) with epidemiological models (e.g., coalescent or birth–death frameworks) to infer transmission history, changes in effective population size, and migration or spread over time, subject to sampling and evolutionary-rate constraints.

Principle

Patterns of genealogical branching and genetic diversity encode information about past transmission and population processes; integrating genetic data with epidemiological models constrains estimates of transmission rates, timing of events and population dynamics beyond what each data type alone can provide.

Demonstration

Illustrative scenario: During an outbreak, time-stamped pathogen sequences sampled from cases are used with a birth–death phylodynamic model to estimate the epidemic growth rate and the likely timing of introduction(s); the genetic tree’s branching times inform plausible transmission tempo when mutation accumulation is sufficient for resolution.

Misapplication

Interpreting a sampled pathogen phylogeny as a definitive transmission tree (i.e., claiming that internal nodes correspond to individual transmission events) without accounting for incomplete sampling, within-host diversity, and model assumptions; the error is conflating a population-level genealogy with exact person-to-person transmission events.

Consequence

When applicable, phylodynamic inference refines estimates of reproduction number trajectories, introduction timing and spatial spread pathways, thereby informing surveillance, outbreak reconstruction and control prioritization. Limitations in sampling density or genetic signal reduce reliability.

Reversal

If the pathogen’s molecular evolution is too slow relative to the epidemic timescale, or if sampling is sparse or strongly biased, the genetic data provide little additional information and phylodynamic inferences become weak or unidentifiable; in such cases traditional epidemiological methods may be more informative.

Boundary

Clearly within: analyses that explicitly couple genetic sequence-derived trees with epidemiological population models to infer parameters over time. Boundary case: phylogeographic or phylotemporal descriptions that use trees descriptively without formal epidemiological coupling. Clearly outside: standalone phylogenetic analyses focused solely on taxonomy or deep evolutionary relationships not intended to infer epidemic dynamics.

Semantic Tension

Tension between the additional resolution genetics can provide and the strong data and model assumptions required (sufficient mutation rate, representative sampling, correct model choice); gaining inference power from sequences comes at the cost of sensitivity to sampling bias and evolutionary model misspecification.

Synthesis

Phylodynamic models link evolutionary and epidemiological timescales to recover aspects of transmission history that are otherwise unidentifiable, but their utility depends critically on sampling design and the magnitude of the molecular clock signal relative to the epidemic.