Unable to load scRNAseq data into R with Read10X

Question

[Rogerio Ribeiro]

Greetings,

I'm working on a single cell datasets, sequenced using 10× chromium single-cell platform, which was deposited here.

I have downloaded the barcodes, features and the count matrix and saved when in the same folder. However, then trying to load them into an R object there is an error:

lung_rawc =  Read10X(data.dir = "/home/rribeiro/scRNAseq_GSE136831/data")

Error in dimnamesGets(x, value) :
  invalid dimnames given for “dgTMatrix” object

Any idea what might be causing this issue?

I have also inspected the downloaded files and everything seems right:

gzip -c -d features.tsv.gz | head

"Ensembl_GeneID"        "HGNC_EnsemblAlt_GeneID"
"ENSG00000000003"       "TSPAN6"
"ENSG00000000005"       "TNMD"
"ENSG00000000419"       "DPM1"
"ENSG00000000457"       "SCYL3"
"ENSG00000000460"       "C1orf112"
"ENSG00000000938"       "FGR"
"ENSG00000000971"       "CFH"
"ENSG00000001036"       "FUCA2"
"ENSG00000001084"       "GCLC"

gzip -c -d features.tsv.gz | wc -l    ##number of genes - 1 due to colnames

45948



gzip -c -d matrix.mtx.gz | head

%%MatrixMarket matrix coordinate integer general
45947 312928 692789348
6 1 2
17 1 2
30 1 1
33 1 1
48 1 1
54 1 1
72 1 1
73 1

gzip -c -d barcodes.tsv.gz | head

001C_AAACCTGCATCGGGTC
001C_AAACCTGTCAACACCA
001C_AAACCTGTCACAGTAC
001C_AAACCTGTCTGTCTAT
001C_AAACGGGAGACTAAGT
001C_AAACGGGAGGCTCATT
001C_AAACGGGAGGGAACGG
001C_AAACGGGCAGTAGAGC
001C_AAACGGGGTATAATGG
001C_AAACGGGGTTTAGGAA

gzip -c -d barcodes.tsv.gz | wc -l  # = to the number of cells reported in the manuscript
312928

Answer 1

[ATpoint]

Feature file should not have a header: https://github.com/satijalab/seurat/blob/master/R/preprocessing.R#L840