Given single-cell data that was obtained across different spatial locations or samples, DIALOGUE treats different types of cells from the same micro/microenvironment or sample as different representations of the same entity. It first uses penalized matrix decomposition to preform penalized canonical variates analysis (CVA) on the average cell-type-specific expression of each niches/samples and identify the initial multicellular programs. Each multicellular program is composed of co-regulated cell-type-specific programs that are defined as a sparse linear combination of genes (Fig. 1a, Methods). It then preforms multilevel (hierarchical) modeling to prune and refines the programs based on the underlining distribution of cell states across and within niches/samples (instead of using only the averages), while controlling for potential confounders.
Package details |
|
---|---|
Maintainer | |
License | Free to use! Please cite our paper (Jerby-Arnon and Regev, 2020) |
Version | 1.0 |
Package repository | View on GitHub |
Installation |
Install the latest version of this package by entering the following in R:
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.