Project Summary  

Timely detection and effective investigation of infectious disease outbreaks, including those caused by antimicrobial-resistant (AMR) pathogens in healthcare settings and viral pathogens with pandemic potential, are essential to public health. Pathogen genomics is increasingly central to these efforts, yet current methods are often limited by reliance on cultured isolates, complete genome assemblies, and assumptions of clonal evolution.

 This project will develop computational methods for outbreak detection and investigation using metagenomic long-read sequencing data obtained directly from clinical samples. It will focus on achieving strain-level resolution in polymicrobial infections, while addressing major analytical challenges including incomplete genome recovery and recombination. A key component of the project will be the extension of structured coalescent-based methods to operate in these settings, enabling accurate inference of transmission direction, within-host diversity, and the presence of unsampled intermediate cases.

 To address the limitations of traditional tree-based phylogenetic methods, the project will explore genealogical frameworks that explicitly model recombination and represent evolutionary history as a sequence of local trees. These approaches will be benchmarked and validated using both real and simulated outbreak datasets, spanning bacterial and viral pathogens. The outcomes will include new methods and open-source tools to support high-resolution, real-time genomic epidemiology, with direct application in hospital infection control and global pandemic preparedness. The project aims to bridge the gap between raw metagenomic data and actionable epidemiological insights.

Potential Supervisors

• Dr Alistair Miles (Head of Epidemiological Informatics for the Kwiatkowski Group, Big Data Institute, University of Oxford)

Skills Recommended

• To be confirmed

University DPhil Courses

• DPhil in Population Health

‍