This repository contains scripts to reproduce all performance evaluations, comparisons, and figures in our paper introducing the diffcyt framework.
The diffcyt R package implements statistical methods for differential discovery analyses in high-dimensional cytometry data (including flow cytometry, mass cytometry or CyTOF, and oligonucleotide-tagged cytometry), based on (i) high-resolution clustering and (ii) empirical Bayes moderated tests adapted from transcriptomics.
A preprint of the paper is available from bioRxiv:
- Weber L. M. et al. (2018), diffcyt: Differential discovery in high-dimensional cytometry via high-resolution clustering, bioRxiv preprint. Available here.
The scripts in this repository are organized according to the benchmarking datasets: 'AML-sim', 'BCR-XL-sim', 'Anti-PD-1', and 'BCR-XL'.
Within each dataset, scripts are organized into sub-directories to:
- prepare data (semi-simulated datasets 'AML-sim' and 'BCR-XL-sim' only)
- run methods
- generate plots
Code comments are included to explain the purpose of each script.
Data files for the benchmarking datasets are available from FlowRepository under accession number FR-FCM-ZYL8.
The diffcyt package is freely available from Bioconductor. The stable release version can be installed using the Bioconductor installer as follows. Note that installation requires R version 3.5.0 or later.
# Install Bioconductor installer from CRAN
install.packages("BiocManager")
# Install 'diffcyt' package from Bioconductor
BiocManager::install("diffcyt")
To run the examples in the package vignette, the HDCytoData and CATALYST packages from Bioconductor are also required.
BiocManager::install("HDCytoData")
BiocManager::install("CATALYST")
For details on the development version of the diffcyt package, see the GitHub page.
For a tutorial and examples of usage, see the Bioconductor package vignette (link also available via the main Bioconductor page for the diffcyt package).
