Cardinal · Day 2 of 2
Case studies on statistics and machine learning with MSI data
Friday, November 6, 2026 · 10:00 a.m.–12:30 p.m. Boston time (EST). Led by Kylie Bemis, Sai Srikanth Lakkimsetty, Ethan Rogers, and Yinyue Zhu.
This session
This session will cover analysis of preprocessed MSI data. First we will cover region of interest (ROI) discovery using Cardinal’s univariate and multivariate segmentation methods, and how these methods can be used for prediction and classification. We will then conclude with a case study of differential abundance between ROIs and experimental conditions.
Session schedule
Held online as a Zoom webinar, 10:00 a.m.–12:30 p.m. Boston time (EST; 15:00–17:30 UTC · 7:00–9:30 a.m. PST · 3:00–5:30 p.m. GMT · 4:00–6:30 p.m. CET): two one-hour lectures with a short break, followed by general discussion.
- 10:00–11:00 Lecture 1 — Methods for spatial segmentation, prediction, and classification of MSI datasets in Cardinal
- 11:00–11:15 Break and questions
- 11:15–12:15 Lecture 2 — Differential analysis of multi-tissue MSI datasets with complex designs in Cardinal
- 12:15–12:30 General discussion
Background
This course covers the basics of data manipulation, visualization, and statistical analysis of mass spectrometry imaging (MSI) experiments using Cardinal and R. We begin by introducing the fundamental ways of interacting with MSI datasets and experimental metadata in Cardinal, followed by preprocessing and visualization of mass spectra and ion images. We then demonstrate the application and interpretation of statistical and machine learning methods for segmentation, classification, and hypothesis testing. Sessions mix lectures with code demonstrations using datasets from public repositories such as PRIDE and MassIVE.
Who it’s for
Biologists, chemists, bioinformaticians, computer scientists, data scientists, statisticians, and engineers who are comfortable with the basics of R and interested in working with MSI data in Cardinal.
Software and installation
To work through the code demonstrations, install the software below. The session used recent versions of R and Bioconductor:
- R (≥ 4.6)
- Bioconductor (current release)
- Cardinal
- CardinalWorkflows (example datasets and workflows)
Please install or update to these versions to run the examples.
- Install R (≥ 4.6) from CRAN, and (recommended) RStudio Desktop. On Windows, also install Rtools.
- Open R or RStudio and run the installation code below in the console.
if (!require("BiocManager", quietly = TRUE))
install.packages("BiocManager")
BiocManager::install(c("Cardinal", "CardinalWorkflows"))
Materials
Presentation slides, code, and datasets were shared with registered participants.
Links and resources
- Cardinal project site
- Cardinal on Bioconductor
- CardinalWorkflows on Bioconductor
- Installing Bioconductor
- Download R (CRAN)
- Download RStudio Desktop
All days in this course
- Day 1 · Thursday, Nov 5 Introduction to Cardinal, preprocessing, and visualization
- Day 2 · Friday, Nov 6 Case studies on statistics and machine learning with MSI data