A Beginner’s Guide to Analyzing and Visualizing Mass Cytometry Data

Abstract

Mass cytometry has revolutionized the study of cellular and phenotypic diversity, significantly expanding the number of phenotypic and functional characteristics that can be measured at the single-cell level. This high-dimensional analysis platform has necessitated the development of new data analysis approaches. Many of these algorithms circumvent traditional approaches used in flow cytometric analysis, fundamentally changing the way these data are analyzed and interpreted. For the beginner, however, the large number of algorithms that have been developed, as well as the lack of consensus on best practices for analyzing these data, raise multiple questions: Which algorithm is the best for analyzing a dataset? How do different algorithms compare? How can one move beyond data visualization to gain new biological insights? In this article, we describe our experiences as recent adopters of mass cytometry. By analyzing a single dataset using five cytometry by time-of-flight analysis platforms (viSNE, SPADE, X-shift, PhenoGraph, and Citrus), we identify important considerations and challenges that users should be aware of when using these different methods and common and unique insights that can be revealed by these different methods. By providing annotated workflow and figures, these analyses present a practical guide for investigators analyzing high-dimensional datasets. In total, these analyses emphasize the benefits of integrating multiple cytometry by time-of-flight analysis algorithms to gain complementary insights into these high-dimensional datasets.

Cyt‐Geist: Current and Future Challenges in Cytometry: Reports of the CYTO 2018 Conference Workshops

“…The tools used in the analysis of high‐dimensional data sets are constantly evolving and improving. While the most common solution is the Cytobank platform, many computational analysis tools are available as open‐source packages in programming languages such as “R” . These tools allow for modification and experimentation of existing analysis approaches but require a knowledge in informatics and the use of R. On the other hand, GUI‐based tools like Cytobank allow a wider base of researchers to harness powerful analytic tools.…”

Rigor and Reproducibility of Cytometry Practices for Immuno‐Oncology: A multifaceted challenge

“…This can be a significant source of variability as it causes experimental noise between sample runs, altering staining patterns. Daily calibration is required to ensure optimal running conditions and minimize problems that may interfere with signal detection and affect instrument accuracy (55). One strategy that may allow for more precise comparisons is to barcode individual samples which can then the pooled for batched acquisition.…”

“…Instrument calibration is another critical aspect to consider. Daily calibration is required to ensure optimal running conditions and minimize problems that may interfere with signal detection and affect instrument accuracy (55). Oxidation due to plasma ionization of isotopes is one such example.…”

Impact of selective CD28 blockade on virus-specific immunity to a murine Epstein-Barr virus homolog

“…28 We conducted CITRUS analysis on a panel of immune checkpoint markers (PD-1, TIM-3, TIGIT, 2B4) to identify clusters that were significantly more or less abundant between the groups. 28 We conducted CITRUS analysis on a panel of immune checkpoint markers (PD-1, TIM-3, TIGIT, 2B4) to identify clusters that were significantly more or less abundant between the groups.…”

“…Traditionally, identifying cellular subpopulations has relied on prior knowledge-driven analysis generates a radial tree that portrays events in a hierarchical manner, with parent clusters giving rise to two or more daughters. 28 We conducted CITRUS analysis on a panel of immune checkpoint markers (PD-1, TIM-3, TIGIT, 2B4) to identify clusters that were significantly more or less abundant between the groups. At 14 dpi, there was a significant decrease in the abundance of cells found within parent cluster 17 654 in both the CTLA-4Ig-and anti-CD28-dAb treated samples as compared to no treatment ( Figure 3A,B), but the abundance of cells in cluster 17 654 was not different between the two treatment groups.…”