Email Address: firstname.lastname@example.org
Other Research Theme Collaborations: SURPH
Background: Since January 2017, I am a Career Development Fellow (Postdoctoral researcher) at the MRC Biostatistics Unit. I am working with Paul Newcombe and Steve Burgess, and my research is on Mendelian randomization. Before joining the BSU, I completed a PhD in Statistics at the University of Warwick. Briefly, my PhD research was on the analysis and presentation of results of (Bayesian) inference on statistical models containing categorical explanatory variables. My supervisor was Professor David Firth. Research Interests: Mendelian randomization uses genetic variants (SNPs) as instrumental variables to assess the existence of a causal relationship between a biomedical risk factor and a (disease) outcome. It is an approach for causal inference in genetic epidemiology and has become quite popular in recent years. During my time at the Unit, I have worked on a number of projects in Mendelian randomisation, mainly concerning:
- The use of variable selection techniques to identify suitable sets of genetic variants to be used in a Mendelian randomisation analysis.
- Quantifying and modelling selection bias in Mendelian randomisation studies.
Selected PapersGkatzionis, A., Burgess S. and Newcombe, P. J. (2018)Bayesian variable selection with a pleiotropic loss for Mendelian randomization
: In preparation
Gkatzionis, A. and Burgess S. (2018)Contextualizing selection bias in Mendelian randomisation: how bad is it likely to be?
International Journal of Epidemiology : To appear
Burgess, S., Zuber, V., Gkatzionis, A. and Foley, C. N. (2018)Modal-based estimation via heterogeneity-penalized weighting: model averaging for consistent and efficient estimation in Mendelian randomization when a plurality of candidate instruments are valid
International Journal of Epidemiology 47(4): 1242-1254
Gkatzionis, A. (2015)Using quasi-densities to summarize and present the posterior distribution of parameter contrasts in statistical models
: PhD Thesis