Simpler is not always better: Phylodynamic misspecification and deep-learning corrections.

The birth-death (BD) phylodynamic estimator tends to underestimate Re when super-spreading, incubation, or contact-tracing are present, risking misleading transmission estimates from time-scaled phylogenies.

• Why it matters: Real outbreaks often include epidemiological features beyond the BD assumptions, yet BD is still widely used because extended models are computationally harder and can introduce numerical issues.
• What they did: Using simulations, the study compared phylodynamic inference of Re and infection duration d under model misspecification, and evaluated deep-learning estimators designed for complex models with added factors.
• The result: Deep-learning-based estimators for complex, epidemiologically realistic models performed well in simulations both when those factors were present and when they were absent, motivating their broader use.

AI summary of the abstract · may contain errors · not medical advice · read the paper.