Modern -omics datasets are very highly dimensional, covering samples (cells, people, bacteria, etc) across thousands to millions of variables (SNPs, genes, proteins, etc). Participants will learn about methods like PCA, MDS, t-SNE, and UMAP, and their applications in investigating data that lives in high dimensions.
We will cover the motivations and ideas behind the methods and go over examples using both artificial and real-world data. By the end of the session, you will be able to code implementations of dimension reduction as well as visualize your data in useful ways and consider approaches for downstream analysis.
The methods covered have implementations in Python and R, and the course will be taught using Python notebooks (with some equivalent code in R provided).
There will be some linear algebra, analysis, and statistics -- knowledge of these topics will be useful for but not critical to participation.
Coffee break will be provided.
This workshop is open to everyone, but space is limited.
Instructor: Alex Diaz-Papkovich
Fee: $50 (seating is limited)
*Reimbursement is available to McGill students after attending the workshop.
To Register: via Eventbrite
For more information on workshops email us at: workshop-micm [at] mcgill.ca (subject: Workshop)