Error with cell annotation using singleR
1
0
Entering edit mode
@bonydekumar-19283
Last seen 19 months ago

Dear all, I created a single cell summarized object from my cell ragner output as follow datadir <- here("sample1", "outs", "filteredfeaturebcmatrix") sce1 <- read10xCounts(datadir) col <- paste("sample1",colData(sce1)$Barcode,sep="") head(col) colnames(sce1) <- col keepfeature <- rowSums(counts(sce1) > 0) > 0 sce1<- sce1[keepfeature, ] Sce1 class: SingleCellExperiment dim: 20065 6560 metadata(1): Samples assays(1): counts rownames(20065): ENSMUSG00000051951 ENSMUSG00000102343 ... ENSMUSG00000063897 ENSMUSG00000095742 rowData names(3): ID Symbol Type colnames(6560): sample1AAACCCAAGAAACTGT-1 sample1AAACCCACAGACCTGC-1 ... sample1TTTGTTGGTCCTGTCT-1 sample1TTTGTTGTCACCGCTT-1 colData names(2): Sample Barcode reducedDimNames(0): spikeNames(0): altExpNames(0): Then i tried to create an object anno as follow anno=MouseRNAseqData() anno class: SummarizedExperiment dim: 21214 358 metadata(0): assays(1): logcounts rownames(21214): Xkr4 Rp1 ... LOC100039574 LOC100039753 rowData names(0): colnames(358): ERR525589Aligned ERR525592Aligned ... SRR1044043Aligned SRR1044044Aligned colData names(2): label.main label.fine Then I normalized sce1 sce1 <- logNormCounts(sce1) sce1 class: SingleCellExperiment dim: 20065 6560 metadata(1): Samples assays(2): counts logcounts rownames(20065): ENSMUSG00000051951 ENSMUSG00000102343 ... ENSMUSG00000063897 ENSMUSG00000095742 rowData names(3): ID Symbol Type colnames(6560): sample1AAACCCAAGAAACTGT-1 sample1AAACCCACAGACCTGC-1 ... sample1TTTGTTGGTCCTGTCT-1 sample1TTTGTTGTCACCGCTT-1 colData names(2): Sample Barcode reducedDimNames(0): spikeNames(0): altExpNames(0): then i tried to annotate cells as follow pred.sce1 <- SingleR(test =sce1, ref = anno, labels = anno$label.main) which is giving following error Error in SingleR(test = sce1, ref = anno, labels = anno\$label.main) : no common genes between 'test' and 'ref'

Both data sets are from mouse.

Any help will be appreciated. Thanks

SingleR • 287 views
1
Entering edit mode
@steve-lianoglou-2771
Last seen 1 hour ago
Denali

Take a look at rownames(anno) vs rownames(sce1).

It looks like the rownames of anno are gene symbols, and your sce1 is using Ensembl gene identifiers (as well you should be). You'll need to convert the rownames of one to the other. There are many ways you can do that ... I'll just point you to one thread you can use to begin your ID conversion journey.

0
Entering edit mode

Thanks a lot. This was indeed helpful.