Introductory lecture on stats methods
Trainer: Ricard Argelaguet
Overview: An integrative analysis of multi-omics data is essential to leverage the power of modern omics technologies and obtain comprehensive insights into the molecular mechanisms underlying a biological process or disease. The heterogeneity and high-dimensionality of the data however often poses challenges for the data integration. This lecture provides an overview of the basic statistical concepts that are important for the joint analysis of multi-modal data. It discusses statistical properties of multi-omics data and the resulting challenges for the data analysis, and provides an overview on different strategies for both supervised and unsupervised integration of multi-omics data. As important examples it covers both the concepts of regression-based analyses to map associations across omics layers as well as matrix factorisation and dimension reduction as a tool for the visualisation and unsupervised integration of omics data.
Learning outcomes
By the end of this session you will be able to:
- Identify statistical challenges and pitfalls in multi-omics integration
- Explore different integration strategies & evaluate the right strategy for your application
- Describe the statistical concepts underlying common integration methods
Materials: