DR.SC

=========================================================================

DR-SC: Joint dimension reduction and spatial clustering for single-cell/spatial transcriptomics data

DR.SC (Method name is DR-SC) is a package for analyzing spatially resolved transcriptomics (SRT) datasets, developed by the Jin Liu’s lab. It is not only computationally efficient and scalable to the sample size increment, but also is capable of choosing the smoothness parameter and the number of clusters as well.

Check out our NAR paper and our Package vignette for a more complete description of the methods and analyses.

DR.SC can be used to analyze experimental dataset from different technologies with different resolutions, for instance:

  • ST plaform
  • 10X Visium platform
  • SeqFISH, MerFISH, etc
  • Slide-seq, Slide-seqV2, etc.
  • Other platforms…

Once DR-SC model is fitted, the package provides functionality for further data exploration, analysis, and visualization. Users can:

  • Identify clusters
  • Extract low-dimensional embeddings
  • Find significant gene markers
  • Visualize clusters and gene expression using spatial coordinates or 2-dim tSNE and UMAP

To further investigate transcriptomic properties, combining the results from DR.SC and other packages, users can:

  • Infer the cell/domain lineages
  • Infer RNA velocity if splicing and unsplicing matrix (can obtained from raw fastq data) are available
  • Detect conditional spatially variational genes
  • Conduct cell-deconvolution

Installation

To install the the packages “DR.SC”, firstly, install the ‘remotes’ package. Besides, “DR.SC” depends on the ‘Rcpp’ and ‘RcppArmadillo’ package, which also requires appropriate setting of Rtools and Xcode for Windows and Mac OS/X, respectively.

# Method 1: Install it from CRAN
install.packages("DR.SC")

# Method 2: Install it from github
install.packages("remotes")
remotes::install_github("feiyoung/DR.SC")

Usage

For usage examples and guided walkthroughs, check the vignettes directory of the repo.

Setup on Linux system

For parallel compuation based on Rcpp on Linux, users require to use the following system command to set the C_stack unlimited in case of R Error: C stack usage is too close to the limit.

ulimit -s unlimited

Demonstration

For an example of typical DR.SC usage, please see our Package vignette for a demonstration and overview of the functions included in DR.SC.

News

DR.SC version 3.0

  • Add the approximated PCA to speed up the computation for initial values; see functions DR.SC and DR.SC_fit.