Colloquium - Faculty Threads
Title: Faculty Threads (Stan Alama, Angelo Canty, Deirdre Haskell)
Speakers: Stan Alama (McMaster), Angelo Canty (McMaster) and Deirdre Haskell (McMaster)
Some Ongoing Research in Mendelian Randomization - Angelo Canty
Mendelian randomization (MR) is a process to estimate a causal link between an exposure and a phenotypic outcome when a randomized clinical trial is not possible. Genetic markers which are associated with the outcome are used as a surrogate for randomization. The method is an application of the instrumental variables technique used in econometrics where the genetic markers play the part of the instrumental variables. The method is quite powerful and has been used successfully in many situations but there are a number of known issues. In ongoing research with my PhD student and I are examining some of these issues and ways to deal with them. Among the topics we are looking are sample overlap, reliable inference for the causal estimates and the impact of the Winner’s Curse. In this short talk I will introduce this technique and describe some of our preliminary findings from simulation studies.
Defects in liquid crystals - Stan Alama
Equilibrium configurations in nematic liquid crystals are described by a problem in the Calculus of Variations which is closely related to the Harmonic Map problem in differential geometry. Notably, liquid crystals exhibit more types of defects (discontinuities in minimizing configurations) than in the geometry problem. I will discuss some recent results on “Saturn Ring”defects, obtained with L. Bronsard, D. Golovaty, and X. Lamy.
Some model theory, analytic functions and valued fields - Deirdre Haskell
I will explain some motivating concepts of model theory in the context of a recent theorem of mine about analytic functions on valued fields.
Date/Time: Friday September 18, 3:30 - 4:30
Join Zoom Meeting
Meeting details to follow