Santiago Cárdenas

Many emerging pathogens exhibit subcritical transmission (i.e. 0<R0<1) when first introduced to novel host species. If left untreated, the pathogen may evolve (as is feared for H5N1) or find a more efficient mode of transmission (as in the 2022 and 2024 global mpox outbreaks). Controlling the incidence of these pathogens while subcritical is crucial to reducing disease burden and minimizing the risk of outbreaks. For emerging zoonoses, cases arise from a mixture of introductions (via spillover) and onward transmission. There are often structured heterogeneities in the focal host population that lead to elevated spillover or transmission rates, due to variable contact patterns or differences in susceptibility or infectiousness. Such heterogeneities offer the potential for efficient targeted control, but the dynamics and control of subcritical pathogens in heterogeneous populations have not been studied. We are developing a two-group branching process model to characterize how risk heterogeneity shapes patterns of incidence for subcritical pathogens. Focusing on the disease metrics of case incidence and rate of large outbreaks, we quantify the burden arising from subcritical disease spread and disentangle how these metrics are driven by spillover versus onward transmission. We then compare the impacts of applying preventative control strategies that target subgroups and present control recommendations, modulated by the system parameters. This study expands the theoretical foundation for understanding how heterogeneous risk behaviors shape zoonotic emergence and motivate targeted control.

Recent mpox outbreaks highlight the need for better understanding of its transmission in spillover-driven settings. Monkeypox virus (MPXV) genome sequencing from mpox cases enables detection of APOBEC3 editing signatures. The distribution of these mutations can yield quantitative insights into the reproduction number (R), the average number of secondary cases in a susceptible population. In this study, we model the accumulation of APOBEC3-type mutations along mpox transmission chains and develop a maximum likelihood inference tool to estimate R from a cross-sectional sample of MPXV genomes. We test the method on simulated data to provide guidance on minimum case counts required for reliable inference and characterize how sampling bias affects parameter estimation. Finally, we test our method on MPXV genomes from recent outbreaks. This tool contributes to the repertoire of quantitative techniques that leverage genomic surveillance data to estimate transmission parameters, here designed for application in endemic mpox settings.