README.md

In JSB-UCLA/scDesign2: scDesign2: A statistical simulator for scRNA-seq data with gene correlation captured

scDesign2

A transparent simulator that generates high-fidelity single-cell gene expression count data with gene correlations captured

Installation

if(!require(devtools)) install.packages("devtools")
library(devtools)

devtools::install_github("JSB-UCLA/scDesign2")

If errors occur due to the curl package, this discussion on stackoverflow could be helpful.

If dependency R packages are updated when installing the devtools R package, try closing and restarting the R session before proceeding.

The scDesign2 R package do not have many dependencies (for a complete list, please refer to the DESCRIPTION file). However, it does depend on the parallel package, where we use the mclapply() function for the parallel model fitting of multiple cell types. Unfortunately, mclapply()'s parallelization can only be executed on Unix-like operating systems but not on Windows. Therefore, the model fitting of multiple cell types will take longer on Windows. We will try to fix this issue in the future.

Tutorials

This tutorial introduces how to use the scDesign2 R package, and can be directly used when trustworthy cell type information is available.

If not, then this tutorial can be read, which introduces how cell clustering can be performed and evaluated before using the scDesign2 method.

Citation

Sun, T., Song, D., Li, W.V. et al. scDesign2: a transparent simulator that generates high-fidelity single-cell gene expression count data with gene correlations captured. Genome Biol 22, 163 (2021). https://doi.org/10.1186/s13059-021-02367-2

JSB-UCLA/scDesign2 documentation built on Nov. 2, 2024, 4:26 a.m.