Lecture
This lecture covers the importance of experimental design in genomic data analysis, focusing on reducing technical variability, avoiding confounding factors, and dealing with batch effects. It discusses the design of experiments, statistical solutions to identify and quantify batch effects, and methods to address these unwanted effects in downstream analyses. The lecture also explores common experimental setups, factorial designs, balanced vs. unbalanced designs, and the analysis of unbalanced data. Linear models for data analysis, including simple regression and multiple regression models, are explained, along with the use of contrasts and design matrices in statistical analysis of genomic data.