Table 3.

Bioinformatic techniques for inferring cell–cell interactions.

Platform	Data Source	Method	Reference
CellTalkDB	scRNA-seq	Manually curated database of ligand–receptor pairs from both human and mouse samples.	[46]
iTalk	scRNA-seq	Identifying and illustrating alterations in intercellular signaling network. R package made to analyze and visualize ligand–receptor pair.	[47]
PyMINeR	scRNA-seq	Python maximal information network exploration resource. Fully automates cell type-specific identification, and pathways as well as in silico detection of autocrine and paracrine signaling networks	[48]
CellChat	scRNA-seq	Open source R package that is able to visualize, analyze, and deduce intercellular communications from a data input. Uses mass action models and differential expression analysis to deduce cell state-specific signaling communications. Also provides visualization outputs to compare intercellular communication methods.	[49]
CellPhoneDB	scRNA-seq	Identifies biologically relevant interacting ligand–receptor pairs. Cells with the same cluster are pooled together as one cell state. Ligand–receptor interactions are derived based on the expression of a receptor of one state and a ligand of the other state.	[50]
Giotto	scRNA-seq	Open source spatial analysis platform that contains two modules, Giotto analyzer and Giotto viewer, which are both independent and fully integrated. Analyzer provides instructions about steps in analyzing single-cell expression data, and the viewer provides an interactive view of the data.	[51]
ICellNET	RNA-seq, scRNA-seq, and microarray	Transcriptomic-based framework that integrates a database of ligand–receptor interactions, communication scores, and connections of cell populations of interest with 31 human reference cell types and three visualization methods.	[52]
SingleCellSignalR	scRNA-seq	Open source R platform. Relies on a database of known ligand–receptor interactions called LRdb.	[53]
CCC Explorer	Transcriptome profiles	Java-based software. Uses a computational model to look at cell–cell communications ranging from ligand–receptor interactions to transcription factors and target genes.	[54]
NicheNet	Gene expression data	Open source R platform. Uses a database of ligand–receptor interactions to identify ligand–receptor interactions that could drive gene expression changes	[55]
SoptSC	RNA-seq	Similarity matrix-based optimization for single-cell data analysis. Uses a cell-to-cell similarity matrix via gene marker identification, lineage reference, clustering, and pseudo-temporal ordering. From this information, it predicts cell–cell communication networks.	[56]
SpaoTSC	scRNA-seq	Spatially optimal transporting of the single cells. The method has two major components: (1) constructing spatial metric for cells from scRNA-seq data and (2) reconstructing the cell–cell communication networks from the data and identifying relationships between genes from intercellular relationships. Uses python.	[57]
scTensor	scRNA-seq	Open source R package. Instead of looking at one-to-one cell–cell interactions, this software focuses on many-to-many cell–cell interactions. scTensor looks at a three-way relationship (hypergraph) between ligand expression, receptor expression, and ligand–receptor pairs.	[58]