Bayesian inference of epidemiological parameters from genomic data

Context:
Genomic data are increasingly being used to understand infectious disease epidemiology. Isolates from a given outbreak are sequenced, and the patterns of genetic variation are used to infer how the isolates within the outbreak are related to each other. Unfortunately, the phylogenetic trees typically used to represent this variation are not directly informative about epidemiological processes, such as the temporal variations in the reproduction number R(t) which represents the average number of secondary infections caused by each infected individual. Accurate estimation of parameters such as the R(t) function are vitally important to understand the effect of control measures on infectious disease outbreaks., and therefore provide an evidence basis for control strategies.

Aims:
The aim of this project is to develop methodology in a Bayesian framework which can incorporate both epidemiological data and genomic sequencing data to produce robust inference of epidemiological parameters. This will be tested in a variety of settings, such as when prevalence data is noisy and when only a subsample of the genomic data is available.

Novelty:
Typically, inference of epidemiological parameters is performed using only epidemiological data (Wallinga and Teunis 2004, Abbott et al. 2020). Recent attempts to use genomic data are based on simplistic epidemiological models (Didelot et al. 2014, Didelot et al. 2017). The aims of this PhD project are to develop modelling and inference methodology which formally incorporates genomic data into a flexible epidemiological framework.

Alignment to the research council's remit:
The aims of the project are to develop new statistical methodology, so this falls under the ESPRC research area "Statistics and applied probability".