Single-cell CRISPR screens provide unprecedented insights into gene regulation and other facets of human genome biology. However, the analysis of these screens poses significant statistical and computational challenges. sceptre (pronounced “scepter”) is a methodology and associated R package for rigorously identifying regulatory relationships in single-cell CRISPR screen experiments. sceptre tests whether a given perturbation is associated with the change in expression of a given gene using the robust, powerful, and intuitive conditional randomization test.
Update June 2022: We have released a beta version of the sceptre Nextflow pipeline. This pipeline greatly facilitates the application of sceptre to large-scale data.
Update March 2022: We are excited to release sceptre version 0.1.0, a major update that significantly improves the speed and ease-of-use of the software.
Installation
You can install the development version of the package from Github with the following command:
install.packages("devtools")
devtools::install_github("katsevich-lab/sceptre")You can browse the source code on Github here. sceptre has been tested in R versions >=4.1 on macOS and Linux systems.
Using the software
sceptre has several interfaces, which you can choose between based on the size of your analysis.
Small or moderately-sized analysis: If you are running an analysis of small or moderate size (i.e., the data fit into memory and you are using a single computer), see the standard sceptre tutorial here.
Large-scale analysis: If you are running a large-scale analysis (i.e., the data do not easily fit into memory or you are using a high-performance cluster or cloud), see the sceptre Nextflow pipeline here. The documentation for the sceptre Nextflow pipeline currently is sparse; please open a Github issue if you are interested in using this pipeline, and we will provide support.
Note: sceptre currently applies to high multiplicity-of-infection (MOI; >5 gRNAs/cell) single-cell CRISPR screen data. sceptre has not yet been carefully vetted in low-MOI settings. We are working on developing such an extension, which we expect to be available in 2022.
References
Please consider starring this repository and citing the following if you find sceptre helpful in your research.
Methods papers
T Barry, X Wang, J Morris, K Roeder, E Katsevich. “SCEPTRE improves calibration and sensitivity in single-cell CRISPR screen analysis.” Genome Biology.
T Barry, E Katsevich, K Roeder. “Exponential family measurement error models for single-cell CRISPR screens.” arXiv preprint.
Application paper
J Morris, Z Daniloski, J Domingo, T Barry, M Ziosi, D Glinos, S Hao, E Mimitou, P Smibert, K Roeder, E Katsevich, T Lappalainen, N Sanjana. “Discovery of target genes and pathways of blood trait loci using pooled CRISPR screens and single cell RNA sequencing.” Preprint available on bioRxiv.
