Devon Kohler

Devon Kohler

Post-doctoral researcher in computational biology, building statistical & causal methods for MS-based proteomics, with experience in RNA-seq and multi-omics integration.

Statistics for Quantitative Mass Spectrometry: Methods and Case Studies

Short Course, May Institute on Computation and Statistics for Mass Spectrometry and Proteomics, 2025

Course Link

This short course introduced statistical experimental design and analysis strategies for quantitative mass spectrometry-based proteomics. Participants learned methods for normalization, missing value imputation, summarization of protein abundances, statistical inference, confidence interval estimation, and differential abundance testing, as well as multivariate analysis for biomarker discovery.

The program combined lectures with hands-on case studies using the MSstats family of packages (MSstats, MSstatsTMT, MSstatsPTM, and MSstatsShiny), with two tracks offered:

  • Track 1 (slower, GUI-first) emphasized foundational statistical concepts and MSstats workflows in both R and the Shiny interface.
  • Track 2 (faster, code-first) focused on advanced R-based workflows, processing of large datasets, and comparisons to alternative statistical methods (e.g., MSqRob, limma, proDA).

Case studies included label-free, TMT, and post-translational modification (PTM) experiments, illustrating how different data processing and modeling choices influence results. Participants left with practical skills to design, analyze, and interpret quantitative proteomics experiments using statistically rigorous methods.