scrnaseq
scrnaseq copied to clipboard
script might crash on Seurat::HTOdemux when some HTOs have very low reads
Today, I realized a problem with HTO demultiplexing in bfx1875. When running the snakemake workflow on our cluster, HTO demultiplexing crashes on one of the samples. It crashes when calling Seurat::HTOdemux. I think it is related to the fact that for this sample only two HTOs were used, but there are 4 HTOs in the feature reference. I did run the demultiplexing workflow separately (in bfx1825/scrnaseq/bfx1875.Hupla17_2.scrnaseq_hto.Rmd). I manually specifying only the 2 used HTOs. Then it works. A solution might be to subset the Seurat object and only use HTOs with high enough total reads.