Use this URL to cite or link to this record in EThOS: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.788926
Title: Integrated analysis of epidemiological and phylogenetic data to elucidate viral transmission dynamics
Author: Li, Lucy Mengqi
ISNI:       0000 0004 8499 3299
Awarding Body: Imperial College London
Current Institution: Imperial College London
Date of Award: 2017
Availability of Full Text:
Access from EThOS:
Access from Institution:
Abstract:
While infectious disease outbreaks are often summarised by population averages such as the reproductive number, variation between individuals in terms of onwards transmissions modulates the degree of unpredictability of an epidemic, and it needs to be accounted for in models of infection control. This heterogeneity among individuals can be quantified by the dispersion parameter k of the offspring distribution, a distribution that defines the number of secondary infections per infected individual. I have developed an inference framework to estimate k and other epidemiological parameters by fitting stochastic transmission models to both incidence time series and the pathogen phylogeny. Applying the framework to simulated data, I found that more accurate, less biased and more precise estimates of the reproductive number and k were obtained by combining epidemiologic and phylogenetic analyses. Accurately estimating k was necessary for unbiased estimates of the reproductive number, but it did not affect the accurate estimation of epidemic start date and the probability of sampling an infection. I further demonstrated that inference was possible in the presence of phylogenetic uncertainty by sampling from the posterior distribution of phylogenies. In addition to methodological contributions, I found that the inclusion of sequences in statistical inference for polio improved the precision of parameter estimates. Based on sequences collected from patients during a poliovirus outbreak, the estimated values of k were high regardless of the data used. On the other hand, the k estimates were low when a transmission model was fit to environmental sequences collected in Pakistan, which is still endemic for wild poliovirus. Furthermore, analysis of environmental sequences was informative of seasonality parameters whereas inference from incidence time series alone was not. This type of analysis using environmental sequences would be useful as polio eradication draws to a close as the number of symptomatic cases approaches zero.
Supervisor: Fraser, Christophe ; Grassly, Nicholas Sponsor: Medical Research Council
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.788926  DOI:
Share: