Skip to contents

COVID-19 PBMC Reference (Healthy) Single-Cell RNA-Seq Dataset

Source: R/dataDocumentation.R
covid_reference_data.Rd

This dataset contains a downsampled reference (healthy donor) subset of the Stephenson et al. (2021) COVID-19 PBMC atlas. It has been preprocessed to include log-normalized counts, an author-provided cell-type column, and PCA results, restricted to 5 shared cell types (including CD14 monocytes) and a gene panel that always includes the Yoshida et al. interferon-response signature.

Usage

covid_reference_data

Format

An object of class SingleCellExperiment with 414 rows and 900 columns.

Source

Stephenson E, et al. (2021), processed and hosted at https://doi.org/10.5281/zenodo.18274942.

Details

This dataset underwent the following steps:

  • Downloads the processed reference (healthy) object from Zenodo (https://doi.org/10.5281/zenodo.18274942), itself derived from the Stephenson et al. (2021) COVID-19 PBMC atlas (CZI CELLxGENE).

  • Restricts to 5 shared cell types (CD14 monocytes, CD4 T, CD8 T, B cells, NK_16hi) using the author_cell_type_merged column.

  • Downsamples to at most 180 cells per cell type.

  • Selects highly variable genes using the function getTopHVGs from the scran package, intersected between reference and query, always including the 25-gene Yoshida et al. (2019) interferon-response signature.

  • Performs Principal Component Analysis (PCA) using the function runPCA from the scater package.

References

Stephenson E, et al. (2021). Single-cell multi-omics analysis of the immune response in COVID-19. Nature Medicine 27:904-916.

See also

Use data("covid_reference_data") to load and access the resulting reference dataset, and data("covid_query_data") for the corresponding query (severe COVID-19) dataset.

Examples

# Load and explore the COVID-19 reference dataset
data("covid_reference_data")