Uncovering the role of epigenetics in modifying disease risk

About the Project

Genome-wide association studies (GWAS) have uncovered many genetic associations linked to health outcomes but uncovering the mechanisms behind these associations remains difficult. DNA methylation (DNAm) is an epigenetic modification process providing insights into the regulatory mechanism of GWAS variants.

DNAm can be measured at over 800,000 DNAm sites genome-wide using microarrays. Variation in DNAm arises from genetic and environmental factors(1). By combining genome-wide genotype data with DNAm data, studies have identified methylation quantitative trait loci (mQTLs) - genetic variants associated with DNAm variation. In a mQTL study of >27,750 human blood samples, 45% of DNAm sites had a genetic basis (2). Using causal inference modelling, most disease-associated variants did not exhibit causal effects on DNAm sites, and conversely, most mQTLs did not have causal effects on disease risk (2,3). These findings suggest that these relationships may depend on environmental or cellular context. We need to generate catalogs of mQTLs that are dependent on environmental factors such as cell-type and study their influences on health outcomes.

Aims

The aim is to understand context-specific gene regulation and to identify causal influences between context-specific mQTLs and health outcomes. The student and supervisors will tailor and finalize research plans in the first months to reflect the student’s interests. Examples of research questions that the student may wish to address are:

1. Identify genetic factors for cell-type interacting DNAm variation in blood
2. Identify causal associations between cell-type specific mQTLs and health outcomes.
3. Validate genetic and causal associations in cell-type specific datasets

Methods

This PhD project uses large-scale DNAm and genetic datasets by collaborating with academic centres that participate in the Genetics of DNA Methylation Consortium (GoDMC, http://www.godmc.org.uk/). GoDMC promotes a federated analysis protocol where scripts developed by the student will be shared with contributing cohorts.

The student will identify cell-type specific mQTLs from whole blood DNAm data. This is achieved by estimating the proportions of the relevant cell-types in blood and then testing these cellular proportions for interactions with genotype. This is similar as in gene-environment interaction models.

To identify whether DNAm changes are causes or consequences of the cardiovascular health outcome, the student will apply Mendelian randomization analysis (4). Mendelian Randomization is a genetic epidemiological approach that uses genetic variants as proxies to interrogate potential causal links between exposure (eg cell counts) and outcome (disease). This studentship will provide cross-disciplinary training in state-of-the-art epigenetic, genetic and causal inference and population health data science analyses.

Supervisors

Dr Josine Min, Dr Eilis Hannon, Dr Genevieve Leyden and Prof Jonathan Mill

Keywords

Population health sciences, genetics, methylation, Mendelian randomization, epigenetics.

University of Bristol, Bristol Medical School

Bristol Medical School is the largest and one of the most diverse Schools in the University of Bristol, with approximately 1100 members of staff, 1350 undergraduate, 250 postgraduate taught and 300 postgraduate doctoral research students. The Head of School is Professor Chrissie Thirlwell. The Medical School has two departments: Population Health Sciences and Translational Health Sciences. The School is a leading centre for research and teaching across these areas. Research in the School is collaborative and multi-disciplinary, with staff coming from a wide range of academic disciplines and clinical specialties.

How to apply for this project

This project will be based in Bristol Medical School – Population Health Sciences in the Faculty of Health and Life Sciences at the University of Bristol. Use this information to search for the relevant programme in our online application system. Applicants should include the names of two supervisors (with their agreement after discussion of the research proposal). This project does not come with funding. If you have secured your own sponsorship or can self-fund this PhD for 4 years please visit our information page here for further information on the department of Population Health Science and how to apply.