The Statistics and Data Science team plays an integral part in the research of all scientific programmes in the MRC Epidemiology Unit.
Statistics work is led by Stephen Sharp, and includes the following:
- Providing timely and quality statistical advice on the design, analysis and reporting of all studies to all scientific researchers in the Unit and their collaborators, championing the requirements of the MRC Epidemiology Unit Standard Operating Procedure on Good Analytical Practice.
- Keeping abreast of new methods and trends in statistical thinking, and communicating them to scientists within the Unit.
- Developing and maintaining a library of exemplar code and user-friendly commands which are available to enable Unit researchers to apply advanced methods (e.g. spline regression, multiple imputation, weighted Cox regression for case-cohort studies, genomewide association analyses) efficiently and correctly.
- Playing a leading role in the design, analysis and reporting of all clinical trials undertaken by the Cambridge Epidemiology and Trials Unit.
- Leading and delivering analyses and reporting for the National Diet and Nutrition Survey rolling programme on a full cost-recovery basis.
- Performing quality control, imputation and analysis for big data from genomewide and omics platforms.
- Collaborating with statisticians and epidemiologists from other organisations (e.g. the MRC Biostatistics Unit and the University of Cambridge Department of Public Health and Primary Care) on statistical issues of mutual interest and relevance to Unit research.
- Teaching on the University of Cambridge MPhil courses in Epidemiology and Public Health.
- Providing statistical reviews for papers submitted to medical and epidemiological journals.
Data science work is led by Tom Bishop, and includes the following:
- Developing and promoting federated meta-analysis, which enables cross-cohort analyses without physically pooling the data from each study.
- Supporting the application of other novel areas of data acquisition and processing, including web-scraping techniques and use of deep neural networks.
- Collaborating with external experts (e.g. from the University of Cambridge Department of Applied Mathematics and Theoretical Physics, Department of Computer Science and Technology, and Health Data Research UK) on data science issues of mutual interest and relevance to Unit research.