Research Fellow in Health and Life Sciences

About the role

We are seeking to appoint a highly committed, motivated, and creative scientist to join the Northumbria University bioinformatics research group. The research will develop new data science methods that leverage the large amounts of public data from UKBiobank, dbGAP, and internal Whole Sequence data and use state-of-the-art AI/ML technologies, such as transfer learning, LASSO regression penalty, logistic regression classifier model, supervised mixture elliptical copula model combining them with Bayesian hierarchical. 

The role holder will deliver an integrative 3-step Predictive framework (into a web application): you will firstly model Polygenic risk Scores (PRS) according to individuals’ heritable risk, population genetics, environment-modified, and ancestry-specific components of risk in a Bayesian hierarchical model using priors defined by LASSO regression penalty functions. Secondly, you will refine the Disease Risk Prediction according to the computed individuals’ PRS, clinical parameters, family history of the disease, genetic ancestry, lifestyle, and socio-demographics in a logistic regression classifier model. Thirdly, you will model a supervised mixture elliptical copula model to stratify individuals into optimal groups according to the predicted risk.

The researcher will evaluate the portability and transferability of the developed Predictive framework by leveraging large Genome-wide Association Studies (GWAS) data of prostate cancer, Cardiometabolic traits from UKBiobank, dbGAP, as well as joint replacement patient data. The developed predictive framework will be deployed into a web application linked to Cloud Computing resources.

Further information is available in the job description.

This role is fixed term until November 2026.

Due to the timescales associated with the project and the need to have the appointable candidate in post by 15th December 2024, it will not be possible to consider candidates who do not have a current right to work in the UK.

About the team

The role-holder will be based at Northumbria University, City Campus, Newcastle Upon Tyne, Faculty of Health and Life Sciences, Department of Applied Science. You will closely work with the Principal Investigator (Professor Emile R. Chimusa). Travelling and attending subject specialist meetings and national/international conferences will be required. The nature of the post is such that the role-holder is expected to work and attend at the University’s premises such hours as are reasonably necessary for the effective discharge of the duties of the role, within a normal working week of 37 hours.  The department has strength in life science research, state-of-the-art multi-omics laboratories that generate data-intensive biological resources, and a strong track record of public and industrial funding (www.northumbria.ac.uk/about-us/academic-departments/applied-sciences).

About you

To be successful in the role you will have:

• PhD or soon to be PhD qualified (or equivalent experience) in relevant discipline of bioinformatics, computational biology, genomics data science, statistics genetics, or biological computer science related to genomics/GWAS  
• Good understanding of polygenic risk scores (PRS) analysis, and AI/ML/Bayesian models. 
• Excellent critical programming in Python and R,  and large-scale genomic data analytic skills.
• Ability to develop web applications, and databases (e.g. in PostgreSQL) and handle DevCloud. 
• Experience of working in multidisciplinary teams.
• Authorship in peer-reviewed literature.

There must be a willingness to be flexible with working patterns in line with the demands of the project and to travel if necessary.

Further information is available in the person specification.

If you would like an informal discussion about the role, please contact emile.chimusa@northumbria.ac.uk

Advert information