Using Super Learning to Predict HIV-1 Drug Resistance by Samuel Liu '23

Using Super Learning to Predict HIV-1 Drug Resistance by Samuel Liu ’23, Statistics Colloquium, Wednesday, March 8, 1:10 – 1:50 pm, North Science Building 114, Wachenheim.

Abstract:  Various algorithms can be used to generate a model based on observed data, but the optimal learner for prediction will vary depending on the underlying data-generating distribution.  In this talk, I will introduce the “super learner”, which is a prediction algorithm that applies a set of base models and uses cross-validation to select between them.  I will present the theory behind the super learner, and illustrate its performance using simulations.  Then we will look at how the super learner applies to a data example, predicting the susceptibility of HIV based on viral genotype.  Specifically, we will predict susceptibility to a specific protease inhibitor, nelfinavir, using a set of database-derived non-polymorphic treatment-selected mutations.