Abstract:
|
Many commonly used techniques and analyses in genomics research are mine-fields of possible missteps, and the literature doesn’t often offer much guidance or clarity. What are good practices for data cleaning and preliminary explorations? Are you controlling for the appropriate factors and including the terms you need in your models? Are you using standard ANOVAs or adopting machine-learning methods, and what can you gain from these? How do you use these models appropriately? What new methods are needed for single-cell RNA-seq samples? In this talk we will cover the basics (and the “intermediates”) of standard expression analysis, new methods for single-cell expression, and advanced modeling with predictive and machine learning methods, with a focus on biomarker discovery through visualization, exploration, and simple but flexible workflows.
|