Description

This course will discuss details of the statistical experimental design of quantitative mass spectrometry-based proteomic experiments, and the analysis of the acquired data with multiple data processing tools in MSstats. The topics include normalization, principles of statistical inference, summarization of protein abundances from multiple spectral features, derivation of confidence intervals for fold changes, testing proteins for differential abundance, and multivariate analysis for discovery of biomarker. The participants will perform hands-on analyses of the example datasets with open-source software R, MSstats, and other packages.

Target audience

    Experimentalists, bioinformaticians, computer scientists, data scientists and statisticians looking to enhance their skills for statistical analysis of MS-based proteomics experiments. It will contain both lectures and practical hands-on exercises. A prerequisite to this course is the course ‘Targeted proteomics with Skyline’ or ‘Proteomics and metabolomics with OpenMS’, and ‘Beginner’s statistics in R’ (or an equivalent prior expertise).

References

Susan Holmes and Wolfgang Huber. Modern Statistics for Modern Biology. Cambridge University Press, 2109. ONLINE version and PAPER version.

‘Points of Significance’ in Nature Methods

Speakers

    Nuno Bandeira, Meena Choi, Ting Huang, Olga Vitek