Single-cell Interpretable Tensor Decomposition (scITD) employs the
Tucker tensor decomposition to extract multicell-type gene expression patterns
that vary across donors/individuals. This tool is geared for use with single-cell
RNA-sequencing datasets consisting of many source donors. The method has a wide
range of potential applications, including the study of inter-individual variation
at the population-level, patient sub-grouping/stratification, and the analysis
of sample-level batch effects. Each "multicellular process" that is extracted
consists of (A) a multi cell type gene loadings matrix and (B) a
corresponding donor scores vector indicating the level at which the corresponding
loadings matrix is expressed in each donor. Additional methods are implemented
to aid in selecting an appropriate number of factors and to evaluate stability
of the decomposition. Additional tools are provided for downstream analysis,
including integration of gene set enrichment analysis and ligand-receptor analysis.
Tucker, L.R. (1966)
Single-Cell Interpretable Tensor Decomposition (scITD) is computational method capable of extracting multicellular gene expression programs that vary across donors or samples. The approach is premised on the idea that higher-level biological processes often involve the coordinated actions and interactions of multiple cell types. Given single-cell expression data from multiple heterogenous samples, scITD aims to detect these joint patterns of dysregulation impacting multiple cell types. This method has a wide range of potential applications, including the study of inter-individual variation at the population-level, patient sub-grouping/stratification, and the analysis of sample-level batch effects. The multicellular information provided by our method allows one to gain a deeper understanding of the ways that cells might be interacting or responding to certain stimuli. To enable such insights, we also provide an integrated suite of downstream data processing tools to transform the scITD output into succinct, yet informative summaries of the data.
The package has several dependencies from Bioconductor. To install these:
if (!require("BiocManager", quietly = TRUE))
install.packages("BiocManager")
BiocManager::install(c("ComplexHeatmap", "edgeR", "sva", "Biobase"))
Then, to install scITD from CRAN:
install.packages('scITD')
To install the latest version of scITD from GitHub:
devtools::install_github("kharchenkolab/scITD")
Follow the walkthrough to learn how to use scITD. The tutorial introduces the standard processing pipeline and applies it to a dataset of PBMC’s from 45 healthy donors.
We also created a tutorial for running ligand-receptor analysis. This uses the same dataset as the main walkthrough.
If you find scITD useful for your publication, please cite: