Table I. Current computational methods applied to mass cytometry data
Common ApplicationAdvantagesLimitationsAvailability
PCAVisualize relationships in multidimensional dataFamiliar and establishedLinear dimensionality-reduction techniqueScripts available for MATLAB, Stata, R, SPSS, and so forth
Single-cell resolution
viSNEVisualize relationships in multidimensional dataNonlinear dimensionality-reduction techniqueDifficult to compare groups of samplesPart of CYT: http://www.c2b2.columbia.edu/danapeerlab/html/cyt-download.html
Single-cell resolutionCytobank
ACCENSEIdentify clusters within multidimensional data without losing single-cell resolutionNonlinear dimensionality-reduction techniqueDifficult to compare groups of samplesAfter installation, plug-and-play GUI
Density-partitioned clusters facilitate visualizing cellular subpopulationshttp://www.cellaccense.com/
SPADEVisualize signaling data on an MST of clustered cellsVisualization of fold change dataLoss of single-cell resolutionCytobank www.cytobank.org
FlowJo
FlowSOMCluster cells and visualize marker expressionComputationally undemandingNo stand-alone applicationR code available on GitHub, script editing required for use
Allows multiparameter visualization on a single MSThttps://github.com/SofieVG/FlowSOM
CitrusIdentify cell populations that are associated with an experimental end pointBuilt-in regression analysis to compare groups of samples and identify clusters that are different between end pointsNeed ∼10+ samples/group for valid comparisonsR code available on GitHub
Easy-to-use GUIhttps://github.com/nolanlab/citrus
After installation, plug-and-play GUI
WanderlustConstruct cellular developmental trajectoriesVisualization of growth, differentiation, and morphogenesis of cellsDevelopmental processes must be linear and nonbranchingPart of CYT: http://www.c2b2.columbia.edu/danapeerlab/html/cyt-download.html