SciLifeLab The Svedberg seminar series, Tove Fall

Monday May 16

Tove Fall

Department of Medical Sciences, Uppsala University

Tove Fall received her degree in Veterinary Medicine (2005) and her PhD (2009) from the Swedish University of Agricultural Sciences. Her PhD thesis subject was to characterize spontaneous diabetes in dogs. After her PhD, she undertook a post-doctoral training (2010-2012) at the Karolinska Institutet working on the genetic epidemiology of obesity and diabetes in humans. She now leads a research group at Uppsala University at the Department of Medical Sciences with the aim to identify risk factors and causes for diabetes onset and progression. The group uses both traditional epidemiology methods and latest -omics techniques in large human data-sets with detailed phenotyping.

Mendelian Randomization – Nature’s own clinical trial?

Type 2 diabetes affects at least 400 million people and is predicted to reach pandemic proportions with >590 million sufferers by 2035 and is a major cause of cardiovascular morbidity and mortality. A multitude of risk factors for type 2 diabetes and its progression to cardiovascular disease are reported, but the causal relationships have been difficult to disentangle as many lifestyle factors are correlated. To correctly identify the specific causes is necessary for effective prevention. Randomized clinical trials are not always possible to carry out for these research questions, and come with a large cost. As a solution, we and others have applied instrumental variable analysis to mimic clinical trials using the random allocation of alleles at conception as our randomization procedure. This concept is known as Mendelian Randomization and has been shown to be able to effectively predict outcomes of large randomized clinical trials to a low cost and no risk for patients. The method can be applied to assess both lifestyle and drug effects. However, there are many pitfalls with Mendelian Randomization and the interpretation of results related to the choice of genetic variants included in the instrument. In this talk, I will explain the concept of Mendelian Randomization, describe the limitations, and illustrate the concept with some of our recent studies in the diabetes and cardiovascular field.

Host: Jessica Nordlund