| Summary: | Classical epidemiological theory uses compartmental susceptible-infected-removed (SIR) models to quantitatively explore epidemics caused by acute, immunizing viruses that sweep through local populations. While the basic theory explains viral-host dynamics in some systems, added model complexity creates a more accurate context for other systems. As a case study, I investigate the system of phocine distemper virus (PDV), a morbillivirus, in harbor seals (Phoca vitulina) in the North Sea. Using this dataset, I extend the classical theory by incorporating host age or stage heterogeneities, geographical heterogeneities, and host movement heterogeneities. First, I investigated age or stage heterogeneities in transmission among the harbor seals in the Dutch 2002 PDV epidemic by creating three models to see which best fit the data. The model with the highest degree of heterogeneity best fit the host stage-structure (p=0.0004). I also estimated the who acquires infection from whom (WAIFW) matrix from detailed incidence data and confirmed with an R0 calculation using next-generation formalism. Next, I addressed geographic heterogeneity in the host population by addressing error inherent in the PDV incidence data in the entire North Sea with a Bayesian framework to estimate the initial population of susceptible individuals (S0), the rate of pathogen transmission (â), and the time series of infected individuals with imperfect binomial reporting to the biweekly incidence time series and used this information to create distance and gravity based models to discriminate how different geographic locations are coupled by infected host movement. Results show that the distance model has a better fit to the seal stranding data, indicating that the distance between harbor seal haulouts drives the spatial spread of PDV. Lastly, I looked at the epidemiological and evolutionary dynamics of the viral family Paramyxoviridae by investigating serially-sampled genomes of measles virus, mumps virus, and canine distemper virus. Using a Bayesian coalescent approach, we estimate viral substitution rates, the time to common ancestry and elements of their demographic history. Strikingly, the mean Time to the Most Recent Common Ancestor (TMRCA) was both similar and very recent among the viruses studied, ranging from only 58 to 91 years (1908 to 1943).
|
|---|