Note: Many of these slides have benefitted, have been inspired or simply taken from wonderful colleagues that have made their own lecture material available to the community. Some of the slides are based on a course I previously co-taught with Jenny Bryan (RStudio), Gabriela Cohen-Freue (Statistics, UBC), and Paul Pavlidis (UBC). Also, Special thanks to: Aaron Quinlan (Applied Comp Genomics lectures), Josh Starmer (StatQuest videos), Manolis Kellis (Deep Learning in Life Sciences lectures), Geoffrey Hinton (many NN and ML lectures).
Lecture PDFs are available here.
Lecture 1: Intro to the course, logistics, and student evaluation.
Lecture 2: Review of basic concepts from ML, Statistics, and the Human Genome
Lecture 3: Human Genome(s) and statistics of Genome-wide Association Studies (GWAS)
Lecture 4: Statistics and concepts in GWAS cont'd
Lecture 5: Confounding and Linear Mixed Models (LMMs)
Lecture 6: LMMs, heritability, and non-linear effects
Lecture 7: Modeling epistasis and non-linearity cont'd
Lecture 8: Confounding issues in association studies and basics of causal inference
Lecture 9: Causality and Instrumental Variable Analysis (IVA)
Lecture 11: Deep neural networks for regulatory genomics (CNNs, RNNs, and optimization)
Lecture 13: Explainable AI techniques for model interpretation
Lecture 15: Unsupervised learning for single cell genomics
Lecture 16: Multi-modal learning in single cell genomics
Lecture 18: Special topic: COVID genomics