I am Alex, a second-year Ph.D. student in QLS. Before coming to McGill, I earned a Bachelor of Mathematics at the University of Waterloo, majoring in statistics and minoring in pure mathematics. After graduating, I worked as a mathematical statistician at Statistics Canada where I developed survey methodology for several projects and worked on modernizing the census with administrative databases. During that time I also earned a Master of Science in probability and statistics at Carleton University, writing a thesis on the mining and analysis of National Hockey League statistics under the supervision of Dr. Shirley Mills. I’m currently carrying out research at the McGill University and Génome Québec Innovation Centre under the supervision of Dr. Simon Gravel.
I’ve always been interested in the history of humanity, from our evolution to how we migrated out of Africa and populated the world. We can learn a lot by looking at our DNA and searching for shared signals between populations. Our DNA can be stored numerically in a variety of ways, but because there’s so much data – a human genome is 3 billion base pairs long! – it can be overwhelming and difficult to separate signal from noise. My main research interest is in using dimension reduction strategies to parse and condense this information, visualize it, and use it to study populations and their histories. Given how fast new data is coming out, and how much more of it we anticipate, these strategies will help us answer questions around a variety of topics like ancestry, population health, and the diversity of our societies.
I’m also curious about questions in other fields of biology, and enjoy experimenting with statistics and machine learning methods on all types of biological data. I have ongoing collaborations at the Montreal Neurological Institute to study the genetic architecture of the brain, and I regularly work with my fellow QLS students to figure out ways to understand data in the –omics world.
A visualization of 488,377 individuals in the UK biobank, coloured in by self-identified ethnic background. The placement of points was determined using dimension reduction on genotype data. (Copyright Alex Diaz-Papkovich. Source)