Table 1.
Automated tools for the analysis of cytometric data. For each tool is reported the function, the statistical platform in which they are available, and a brief description of the main function.
| Function | Software | Availability | Description | Reference |
|---|---|---|---|---|
| Pre-processing | FlowCore | R, Bioconductor | Import, compensate and transform FCS files in R environment | [9] |
| FlowStats | R, Bioconductor | Collection of algorithms to analyze flow cytometry data, including correction of batch effect | [10] | |
| FlowClean | R, Bioconductor FlowJo plugin | Quality control of data set based on compositional analysis | [11] | |
| FlowAI | R, Bioconductor FlowJo plugin | Quality control of data set based on flow rate, signal acquisition and dynamic range | [12] | |
| CATALYST | R, Bioconductor | Collection of algorithms to pre-process cytometric data and to perform data analysis (with FlowSOM clustering and dimensionality reduction) | [13] | |
| CytoNorm | R | Normalized batch effect using control sample and clustering algorithm | [14] | |
| Automated sequential gating | FlowDensity | R, Bioconductor | Provides tools for automated 1-D and 2-D sequential gating | [15] |
| OpenCyto | R, Bioconductor | Facilitates automated 1-D and 2-D gating methods in sequential way to mimic the manual gating | [16] | |
| AutoGate | Standalone software | Performs 2-D sequential gating to obviate the need to draw arbitrary gates to define the subsets in a gating | [17] | |
| cytometree | R | The algorithm relies on the construction of a binary tree, the nodes of which represents cellular populations | [18] | |
| EPP | Standalone software | AutoGate extension. Algorithm that detects the best 2-D gating strategy to identify cellular populations | [19] | |
| Boolean combination gates | flowType | R, Bioconductor | Phenotyping cytometric using multi-dimensional expansion of 1-D partitions | [20] |
| FloReMi | R | Starting from flowType results identifies the populations that best correlates with an external outcome | [21] | |
| RchyOptimyx | R, Bioconductor | Starting from flowType results, constructs a hierarchy of cells selecting the most informative phenotypes for biomarker detection | [22] | |
| Clustering | FlowMeans | R, Bioconductor FlowJo plugin | Automated gating tool based on K-means algorithm | [23] |
| SPADE | R, Matlab, Cytobank, FlowJo plugin | Clustering method based combining density-based sampling with hierarchical clustering | [24] | |
| HDPGMM | Python | Clustering based on hierarchical modeling extensions to the Dirichlet Process Gaussian Mixture Model | [25] | |
| Citrus | Cytobank, R | Identifies cell populations with hierarchical clustering and make prediction with regression model | [26] | |
| FlowSOM | R, Bioconductor FlowJo plugin, Cytobank | Clustering method combining SOM and hierarchical clustering | [27] | |
| X-shift | Standalone software, FlowJo plugin | Clustering based on kNN density estimation and cluster merging according Mahalanobis distances | [28] | |
| flowClust | R, Bioconductor | Model-based clustering using a t-mixture model | [29] | |
| immunoClust | R, Bioconductor | Model-based clustering on individual samples. Includes an additional step to map cluster between samples | [30] | |
| SWIFT | Matlab | Clustering method based on splitting and merging of Gaussian mixture models | [31] | |
| FLOCK | C, Immport | Automated method partitioning of each dimension into bins, followed by merging of dense regions, and density-based clustering | [32] | |
| flowPeaks | R, Bioconductor | Clustering method combining density-based clustering and K-means | [33] | |
| ClusterX | R | Fast clustering by automatic search and find of density peaks | [34] | |
| PhenoGraph | Matlab, Python | Cells are visualized in a graph structure and connected with weighted edge based on neighbor shared by cell. Graph is then partitioned in group of cells sharing similar phenotypes | [35] | |
| Dimensionality reduction | t-SNE | FlowJo plugin | Performs t-SNE in FlowJo, allowing to manually gate region in dimensionality reduced space to compare cell frequency across samples | [36] |
| ACCENSE | Standalone software | Performs dimensionality reduction with t-SNE algorithm, followed by clustering of dimensionality reduced events with K-means or DBSCAN algorithms | [37] | |
| Rtsne | R | Performs t-SNE dimensionality reduction in R environment | [36] | |
| viSNE | Cytobank, Matlab | Visualization tool based on implementation of t-SNE algorithm | [38] | |
| EmbedSOM | R, Bioconductor FlowJo plugin | Dimensionality reduction technique based on SOM | [39] | |
| UMAP | R, Python, FlowJo plugin | Dimensionality reduction technique based on Uniform Manifold Approximation and Projection (UMAP) | [40] | |
| Destiny | R, Bioconductor | Performs dimensionality reduction with diffusion map | [41] | |
| Fit-SNE | R, Matlab, Python, FlowJo plugin | Tool to perform dimensionality reduction using Fast Fourier Transform-accelerated Interpolation-based t-SNE | [42] | |
| Trajectory inference | Wanderlust | Matlab | Trajectory inference method based on kNN graph: Developed to identify linear transitions | [43] |
| Wishbone | Matlab, Python | Evolution of Wanderlust, it can identify bifurcation in the trajectories | [44] | |
| Monocle | R, Bioconductor | Identification of bifurcated trajectory based on MST | [45] | |
| PHATE | Matlab, Python | Identification of trajectory preserving continual progressions, branches and clusters | [46] |
R, package or code working on R; Bioconductor, R package available on Bioconductor repository [47]; Python, code or library written in Python language; Matlab, code or software based on Matlab language; C, code based on C programming language; FlowJo plugin, downloadable tools to expand FlowJo functionality [48]; Cytobank, online platform for single-cell analysis [49]; ImmPort, immunology database and analysis portal [50].