Event

Identifying causal risk factors for disease is of central importance in health research. However, when using observational data, causal inference is difficult due to the potential presence of unmeasured confounding. In this talk I will discuss how genetic data can be used to study questions on the causes of disease outcomes, focusing on the Mendelian randomization approach. Mendelian randomization is an increasingly popular tool in genetic epidemiology which uses genetic variants as instrumental variables. The rapid rise in the availability of datasets summarising associations between genetic variation across the human genome and an enormous variety of traits has led to great potential for the technique to contribute to the evidence base in health research. However, in order to make valid causal inference under the Mendelian randomization paradigm, a number of assumptions are required to be made which are not always easy to justify. I will discuss some of these limitations and outline methodological advances which allow for Mendelian randomization analyses to be performed when these assumptions are not met.

Andrew is a Lecturer in Biostatistics in the Sydney School of Public Health at the University of Sydney. Prior to joining the University of Sydney, he was a Research Associate in the MRC Biostatistics Unit at the University of Cambridge, working primarily on methods development for Mendelian randomization. He has previously worked at Macquarie University, where he also completed his PhD in Statistics. Andrew’s main research interests span the fields of causal inference, statistical genetics, and time series analysis, with the overall aim of developing statistical methods which can be used to gain insight into the causes and mechanisms of disease.

This presentation is being held via Zoom and is being recorded.

Please click this URL to join 

Or, go to https://monash.zoom.us/join and enter

Meeting ID: 857 1872 3714

Passcode: 551728