DIALOGUE is a dimensionality reduction method that uses cross-cell-type associations to identify multicellular programs (MCPs) and map the cell transcriptome as a function of its environment. Given single-cell data, it combines penalized matrix decomposition with multilevel modeling to identify generalizable MCPs and examines their association with specific phenotypes of interest.
To install DIALOGUE you can either use devtools::install_github(repo = "https://github.com/livnatje/DIALOGUE") or just download its R package and use devtools::install("DIALOGUE")
The input consistes of single-cell transcriptomes of different cell types, usually together with a more compact representation (e.g., PCs). The output will be multicellular programs (MCPs) of co-regulated genes across the different cell types, their expression across the cells, and association with specific phenotype(s) of interest. Each MCP consists of multiple cell-type-specific gene subsets.
For specific cell-cell "interactions" you can run the pairwise version, using the data of two cell types of interest as input. DIALOGUE can also identify MCPs that span multiple cell types (as we show in our pre-print Jerby-Arnon and Regev bioRxiv 2020).
See the tutorial for more details.
- R (tested in R version 3.4.0).
- R libraries: lme4, lmerTest, PMA, plyr, matrixStats, psych, stringi, RColorBrewer, unikn, reshape2, ggplot2, grid, beanplot, parallel
Want to hear more about DIALOGUE and other approaches to study multicellular biology? Checkout our seminar at BCH Digital Science TV