FindAPhD Weekly PhD Newsletter | JOIN NOW FindAPhD Weekly PhD Newsletter | JOIN NOW

Using machine learning to facilitate rapid and efficient responses to gastrointestinal disease outbreaks


   Bristol Veterinary School

This project is no longer listed on FindAPhD.com and may not be available.

Click here to search FindAPhD.com for PhD studentship opportunities
  Prof Kristen Reyher  No more applications being accepted  Competition Funded PhD Project (Students Worldwide)

About the Project

The project:

Illness caused by the consumption of contaminated food is a major threat to public health and requires significant resources to identify the source of infections. However, due to the interconnected nature of the global food supply chain, source identification is complex and labour-intensive. Spearheaded by the UK Health Security Agency (UKHSA), public health agencies have begun to routinely utilise whole genome sequencing (WGS) for pathogen surveillance. WGS provides high-resolution information on the genetic relatedness between disease isolates, allowing for confident identification of clusters of infections arising from common sources.

This project will develop cutting-edge machine learning (ML) tools for the prediction of foodborne disease outbreaks. The student will develop a reinforcement learning (RL)-based decision support tool trained on genomic surveillance data to identify whether we can predict the occurrence of future outbreaks of gastrointestinal disease and determine the optimal time for epidemiologists to intervene. 

The project will achieve three primary objectives:

a) Establish the costs of outbreaks and interventions

Whilst interning with the UKHSA, the student will co-produce knowledge with UKHSA stakeholders to identify costs associated with early, late and no intervention during outbreaks. 

b) Build and train a RL-framework for intervention prediction

Use ~55,000 clinical cases of Salmonella enterica and ~8000 Escherichia coli collected by UKHSA to construct a model parameterized using the outcomes from a) and trained on a combination of genetic and clinical metadata. The model will incorporate recent advances in ML to effectively generate dynamic intervention decisions. 

c) Generating human-readable and understandable outputs from ML

Compare model outcomes from b) to those used by UKHSA epidemiologists to assess the impact of RL decision support on future outbreaks. The model will be packaged to produce human-readable outputs for use by non-experts using explainable ML methods.

The studentship will be hosted in the University of Bristol, which is ranked 5th in the UK for research. The student will be supported by a comprehensive multi-disciplinary supervisory team who are embedded in a number of rich active research communities in the areas of genome bioinformatics, epidemiology, machine learning, AMR and public health in the Universities of Bristol (Prof Kristen Reyher, Prof Andrew Dowsey, Dr Sion Bayliss) and Bath (Dr Lauren Cowley). 

How to apply:

Via the online form until 5 pm on Wednesday, 2nd November 2022

Additional guidance for students can be found here. Please contact [Email Address Removed] or [Email Address Removed] for more information.

Candidate requirements:  

The project would suit an applicant with a strong first degree or masters involving bioinformatics or computational biology. The project can be tailored to the student: those with a mathematical background open to learning skills in bioinformatics and ML or those with a biological/biomedical background and experience in the areas of basic programming, data science or ML are encouraged to apply. 

Please visit https://gw4biomed.ac.uk/doctoral-students/ for more information. Standard University of Bristol eligibility rules apply. Please visit PhD Veterinary Sciences for more information.

Contacts: [Email Address Removed]


Funding Notes

Funding is provided through the GW4 Biomed DTP programme - https://gw4biomed.ac.uk/.
Search Suggestions
Search suggestions

Based on your current searches we recommend the following search filters.

PhD saved successfully
View saved PhDs