Research Assistant (Computational Data Analyst

Research Assistant (Computational Data Analyst)

Job number MED
05897 Faculties Faculty of Medicine Departments Department of Metabolism, Digestion and Reproduction Salary or Salary range £45,399 - £48,876 per annum Location/campus Hammersmith Campus
- Hybrid Contract type work pattern Full time
- Fixed term Posting End Date 19 Aug 2026

We are seeking an experienced Research Assistant with strong data analysis skills to join an exciting Breakthrough T1D-funded research programme led by the Misra Group at Imperial College London. The programme uses large-scale real-world datasets to investigate metabolic dysfunction in type 1 diabetes and its relationship with clinical outcomes, and to develop and validate prediction models that support more individualised care.

Research will focus on near-term complications, metabolic dysfunction, the impact of GLP-1 receptor agonists and other treatments, and long-term cardiovascular risk, combining comparative analyses with risk prediction tools designed to inform clinical practice and policy. The post is based within the Misra Group in the Department of Metabolism, Digestion and Reproduction, a multidisciplinary clinical diabetes research team integrating population-based research, advanced data science, genetics and clinical studies to understand heterogeneity in metabolic health and treatment response across diverse populations with type 1 and type 2 diabetes.

The group brings together enthusiastic and motivated clinical and basic scientists, PhD students, postdoctoral researchers, data analysts and research nurses in a collaborative research environment. For further information, please contact

Responsibilities will include assembling, managing and analysing large real-world datasets to address defined research questions and investigate trends in outcomes among people with diabetes. Advanced epidemiological and statistical methods will be applied, including causal inference approaches such as target trial emulation, active-comparator new-user designs, self-controlled case series and case-crossover studies, where appropriate. The role will also involve developing and validating clinical risk prediction models, including feature engineering, management of missing data, model selection, calibration, discrimination, decision-curve analysis and assessment of clinical utility, alongside internal, external and temporal validation and evaluation of transportability across datasets and population subgroups.

Clinical questions will be translated into robust data pipelines and reproducible analytical code incorporating version control, testing and peer review. Working closely with the PI and multidisciplinary collaborators, the successful candidate will contribute to the intellectual direction of the programme, communicate and interpret analytical methods and findings, and manage day-to-day project activities, deliverables and timelines to ensure successful delivery of the programme’s objectives.

We are looking for an experienced and motivated researcher with a strong background in biostatistics, epidemiology, bioinformatics, computational science or a related quantitative discipline. You will have experience analysing large-scale population health data and applying advanced statistical methods to address clinically important questions, with particular expertise in pharmacoepidemiology, statistical modelling and the development of clinical risk prediction tools. You should be confident working with large and complex datasets, using statistical programming languages such as R, Python, Stata or SAS, and have a strong understanding of data quality, reproducibility and robust analytical practice.

Experience of SQL, cloud-based or high-performance computing environments, and Bayesian methods would also be valuable. Beyond technical expertise, we are looking for someone who is intellectually curious, enjoys solving complex problems and wants to contribute to the scientific direction of an ambitious research programme. You should be able to work both independently and collaboratively, communicate complex methods and…