Module: Advanced Bayesian statistics for genomics

CSB1021H/F, Teaching Section LEC 0150

Coordinator: Professor Guillaume Filion

Offered: Winter 2024 session, in January and February for a total of six meetings.

Weight: One module (0.25 FCE)

Time: TBA

Location: Remote

Enrolment: Limited to 10 students

Description:

This course is an in-depth introduction to Bayesian inference with the variational inference method using the programming language Pyro. It covers the basic theory of variational inference and consists of practical applications to concrete problems (vaccine efficiency, mutation rates, single-cell transcriptomics). The content consists of video lectures explaining how to use Pyro and Pytorch, and in-person lectures with laptop computers to work on practical applications. This course requires some familiarity with statistics and working knowledge of the Python programming language.

Evaluation:

The evaluation consists of a take-home exam to solve problems similar to those addressed in class. 

Pre-requisites for module: No particular classes are required, but the students should be familiar with statistics and standard distributions (Poisson, binomial normal, etc.) and have working knowledge of the Python programming language.

Reading materials: None

Website: TBA

Last updated on August 10th, 2023