Error in Integrating multiomics RNA and ATAC data after removing one cell-type from RNAseq data

[elhaam]

Hi there, thanks for this helpful package!

I am integrating single-cell RNA-seq and ATAC-seq data where one cluster is missing in one dataset, so the sizes of the two datasets are different. I can run LIGER when the cells match in RNAseq and ATACseq data. But after removing one cluster from RNAseq data, I get the following error in LIGER.

CLICK to see my output and the ERROR

Warning message: `legend.margin` must be specified using `margin()`. For the old behavior use legend.spacing Warning message: In mde(x) : NAs introduced by coercion Removing 53637 genes not expressing in atac. Warning in selectGenes(int.liger, datasets.use = 2) : No genes were selected; lower var.thresh values or choose 'union' for combine parameter

Error in (function (cl, name, valueClass) :
assignment of an object of class “NULL” is not valid for @‘var.genes’ in an object of class “liger”; is(value, "vector")is not TRUE
Calls: selectGenes ->
Execution halted

I followed the tutorials and can run it without errors when the number of cells matches between RNA-seq (1047 cells) and ATAC-seq (1047 cells) data.

Is the error happening because of the size difference between the number of cells in RNA-seq (707 cells) and ATAC-seq (1047 cells) matrices? Could you help me with this?

Thanks!

[cgao90]

Hi @elhaam ,

LIGER for iNMF requires the genes to be the same for input datasets (genes by cells matrices). It is likely that the inputs were cells by genes matrices in your case. Could you please check on that?

[elhaam]

Hi @cgao90 Thank you very much for getting back to me. Yes, I checked on that and my RNA and ATAC are both genes by cells matrices. Do you possibly know how I can fix this issue?

Thank you very much in advance!

[skpalan]

Hi, what threshold you are using for selectGenes ? Can you try a lower var.thresh?