Are you an EPFL student looking for a semester project?
Work with us on data science and visualisation projects, and deploy your project as an app on top of Graph Search.
This lecture by Qingyuan Zhao explores the history and concepts of Mendelian randomization (MR), a method using genetic variation to study causal effects. Zhao discusses the old ideas from Mendel to Fisher, the surge in MR popularity, challenges like pleiotropy, and recent methods like MR.RAPS. The lecture delves into the IV diagram, core assumptions, and the RAPS approach to handle pleiotropy. Zhao also covers within-family MR, genetic trio studies, and ongoing work on adjustment sets and randomization tests, emphasizing the importance of MR in causal inference despite challenges like pleiotropy and computation.