An optimal framework for disease risk prediction and stratification for multi-ethnic populations (ref: SF22/HLS/APP/Chimusa)

The exceptional polygenicity of human traits makes unravelling mechanisms from disease risk prediction models daunting. The development of polygenic risk scores (PRS) methods, their evaluation and clinical utility have been explored almost in European ancestry Genome-Wide Association Studies (GWAS) data sets. This makes current PRS predictive power substantially lower when computed in multi-ethnic populations. This problem has been widely reported in recent years, consequently, the etiological insights and clinical utility provided by PRS from existing tools may have little relevance to multi-ethnic populations. In addition, this raises the question as to how the clinical utility of these methods can be made equitable across multi-ethnic populations and, specifically, how to accurately predict disease risk in diverse populations.

There is a critical need to develop novel methods for optimizing the predictive power of disease risk in diverse populations. Harnessing, the power of data-driven artificial intelligence coupled with a family history of the disease, environmental exposures, and genetic effects in designing disease risk prediction models has the potential to achieve unbiased and powerful estimates of risk prediction across diverse populations.

In brief, given the availability of multi-ethnic populations datasets

from UK-biobank (https://www.ukbiobank.ac.uk/) and NIH dbGAP (https://www.ncbi.nlm.nih.gov/gap/) sources, this project targets the unique and complex impact that the environment, family history of the disease, genetics and clinical covariates can have on disease risk in multi-ethnic populations to enable generalizability and transferability of disease risk prediction. This PhD project aims at developing a framework for optimally predicting individuals’ risk of disease conditional on their PRS, family history of the disease, environmental exposures, and ancestry effects has the potential to achieve unbiased and powerful estimates of risk prediction across diverse populations. This overall aim will be accomplished by following three specific lines of investigation: 1) Build a simulation framework to benchmark current PRS tools and guidance on the choice of PRS tool; 2) Develop an integrative admixture-sensitive disease risk prediction approach to address cross-population transferability and generalizability of risk prediction and stratification model; 3) Apply and benchmark the developed risk prediction and stratification tools on simulated data and data from UK-biobank and NIH dbGAP source on cardiometabolic diseases and a spectrum of cancers.

The project will begin by building a simulation framework tool for assessing the predictive power of current PRS tools in relation to the genetic distance between founding populations, miss-specification of ancestral populations and several admixture events, and will be improved by modelling these factors. Robust estimation of SNP, family history of the disease, environmental exposure, clinical variables, and ancestry effects will optimize the developed risk prediction and stratification approaches in multi-ethnic populations through a full hierarchical Bayesian model and LASSO regression penalty functions. A substantial evaluation of developed tools will be done using simulated and real multiples genomics data sets.

Eligibility and How to Apply:

Please note eligibility requirement:

•      Academic excellence of the proposed student i.e. 2:1 (or equivalent GPA from non- UK universities [preference for 1st class honours]); or a Masters (preference for Merit or above)

•      Appropriate IELTS score, if required

For further details of how to apply, entry requirements and the application form, see https://www.northumbria.ac.uk/research/postgraduate-research-degrees/how-to-apply/

Please note: All applications must include a covering letter (up to 1000 words maximum) including why you are interested in this PhD, a summary of the relevant experience you can bring to this project and of your understanding of this subject area with relevant references (beyond the information already provided in the advert). Applications that do not include the advert reference (e.g. SF22/…) will not be considered.

Deadline for applications: Ongoing

Start Date: 1^st October and 1^st March are the standard cohort start dates each year.

Northumbria University takes pride in, and values, the quality and diversity of our staff and students. We welcome applications from all members of the community.

Informal enquiries to Prof. Emile R. Chimusa ([Email Address Removed])