Supervisors: Dr Dylan Neil Clements, Professor Mark Bronsvoort
Project Description:
The Dogslife Project is the first large-scale population-based epidemiological study documenting the incidence and prevalence of diseases in dogs over their lifecourse, and the environmental influences that affect them. It is a unique and unparalleled big data resource, as the information is collected directly from pet owners, and has been shown to be both valid and robust (1). To date there are 8,000 dogs registered with the project, with data collected on over 30,000 illness episodes, and with dogs followed for up to nine years of their life. A multitude of illness phenotypes have been reported in the cohort (including non-infectious diseases (2), infectious diseases, and diseases with a known genetic components). The project has already identified environmental risks which are associated with canine phenotypes, such as weight gain in early adulthood, gastrointestinal disease and limber tail. Genetic risks for diseases have also been mined through the project (3).
This interdisciplinary PhD project will evaluate the rich and diverse dataset collated by Dogslife to provide a doctoral training in quantitative epidemiology and genomics. During the 4-year studentship the successful candidate will evaluate existing and prospectively collected environmental, host genetic and microbial datasets for aging phenotypes, and the interactions between them. In particular, a focus will be made to identify the early-life events which are associated with the development of high morbidity diseases and cumulative disease burdens later in life. Multiple risk variables such as duration and severity of the clinical signs, co-morbidities and preventative healthcare status will be modelled. The results will quantify lifestyle, environmental and genetic risks for common diseases and generalised disease burdens, and where appropriate be used to develop guidelines for owners to guide how they can reduce the risk of disease in their pet.
A range of exploratory and analytical statistical/epidemiological techniques will be taught to the student, to enable the identification of single and composite risk factors for specific disease syndromes, and overall disease burden. During the studentship the candidate will learn and develop a diverse range of skills, including R programming, command line tools, SQL database analysis, survey and experimental design, and the collation and analysis of Google Trends and Google Analytics datasets. Analysis will include the use of multivariate regression models, machine learning, data simulations, Bayesian techniques, and complex data visualisation and mapping. The student can also engage in wet-laboratory work, through sample extraction and preparation, leading to genome wide SNP chip and whole genome sequencing data acquisition, and subsequent analysis. Furthermore the project affords a successful candidate opportunities to present their work to a variety of different audiences, including the general public. This studentship provides the candidate with a fantastic opportunity to develop a bespoke, multidisciplinary research profile with impactful results, on a project that captures the public imagination.
References:
(1) Pugh et al (2015) Journal of Medical Internet Research, doi: 10.2196/jmir.3530
(2) Pugh et al (2017) Preventative Veterinary Medicine, 10.1016/j.prevetmed.2017.02.014
(3) Raffan et al (2016) Cell Metabolism, 10.1016/j.cmet.2016.04.012
If you wish to apply for this project, please check this link and send your application to this email.
