Skip to contents

Handling single-cell RNA-seq data in GEOquery

Sean Davis

2025-08-29

Source: vignettes/single-cell.qmd 
library(GEOquery)
library(SingleCellExperiment)
library(DropletUtils)

Single cell searching 

res = searchGEO("single cell mtx matrix")
head(res)

Mtx files 

poss= getGEOSuppFiles('GSE248214', fetch_files = FALSE)
poss

Multiple h5ad files 

library(GEOquery)
s = getGEOSuppFiles('GSE161228', fetch_files = FALSE)
s

Single h5ad file

SAMPLES

mtx file 

s = getGEOSuppFiles('GSM7908437')
s
sample_prefix = sub('barcodes.tsv.gz','', grep('barcodes.tsv.gz', s$filepath, value = TRUE))
gsm_sce <- DropletUtils::read10xCounts(samples = sample_prefix, type = 'prefix')
gsm_sce
head(colData(gsm_sce))

Tar of mtx files from Series record 

s = getGEOSuppFiles('GSE248214')
s

From here, untar and then wrap to get prefixes

tar_filename = rownames(s)[1]
exdir = tempdir()
untar(tar_filename, exdir=exdir)
sample_prefixes = sub('matrix.mtx.gz','',dir(exdir,pattern='*matrix.mtx.gz',full.names = TRUE))
gse_sce = DropletUtils::read10xCounts(samples = sample_prefixes, type='prefix')
gse_sce

Mix of types in a single GSE

10x h5 file 

s = getGEOSuppFiles('GSM8775062')
sce = DropletUtils::read10xCounts(s$filepath)
sce

10x matrix mtx files 

s = getGEOSuppFiles('GSM8775066')
sce2 = DropletUtils::read10xCounts(samples = '/Users/davsean/Documents/git/GEOquery/vignettes/GSM8775066/GSM8775066_vivo_day7_1_', type='prefix')
sce2