In drisso/scRNAseq: Collection of Public Single-Cell RNA-Seq Datasets

library(BiocStyle)
knitr::opts_chunk$set(error=FALSE, message=FALSE, warning=FALSE)

Downloading the data

We obtain a single-cell RNA sequencing dataset of the mouse haematopoietic stem cells from @paul2015transcriptional. Counts for endogenous genes and spike-in transcripts are available from the Gene Expression Omnibus using the accession number GSE72857. We download and cache them using the r Biocpkg("BiocFileCache") package.

library(BiocFileCache)
bfc <- BiocFileCache("raw_data", ask = FALSE)
base.url <- file.path("ftp://ftp.ncbi.nlm.nih.gov/geo",
    "series/GSE72nnn/GSE72857/suppl")
fname.A <- bfcrpath(bfc, file.path(base.url, "GSE72857_umitab.txt.gz"))

We read this into memory as a sparse matrix.

library(scater)
counts <- readSparseCounts(fname.A, quote='"')
dim(counts)

Downloading the metadata

We pull down the metadata from GEO as well.

meta.A <- bfcrpath(bfc, file.path(base.url, 
    "GSE72857_experimental_design.txt.gz"))
meta <- read.delim(meta.A, skip=19, header=TRUE, stringsAsFactors=FALSE)
meta <- DataFrame(meta)
meta

We check that the cell names match up with the matrix.

m <- match(colnames(counts), meta$Well_ID)
stopifnot(all(!is.na(m)))
meta <- meta[m,]

Saving to file

We now save all of the relevant components to file for upload to r Biocpkg("ExperimentHub").

path <- file.path("scRNAseq", "paul-hsc", "2.2.0")
dir.create(path, showWarnings=FALSE, recursive=TRUE)
saveRDS(counts, file=file.path(path, "counts.rds"))
saveRDS(meta, file=file.path(path, "coldata.rds"))