Hi,
I am doing some scRNAseq analysis and have run into an error when in the normalization step. I have finished QC ending with stats <- perCellQCMetrics(sce, subsets=list(Mito=which(location=="MT"))) high.mito <- isOutlier(stats$subsets_Mito_percent, type="higher") sce <- sce[,!high.mito] This upper part ran successfully.
Now wanted to perform normalization then cluster cells using scran but ran into problem.
library(scran) set.seed(1000) clusters <- quickCluster(sce)
Error in base::colSums(x, na.rm = na.rm, dims = dims, ...) : 'x' must be an array of at least two dimensions
From what I have read in various forums, it seems the colSums() is applied to matrices with more than one column. Exploring the header of sce indicate that it has 6 columns.
DataFrame with 6 rows and 6 columns sum detected subsets_Mito_sum subsets_Mito_detected subsets_Mito_percent total <numeric> <integer> <numeric> <integer> <numeric> <numeric> AAACCTGAGAAGGCCT-1 1738 748 111 11 6.38665 1738 AAACCTGAGACAGACC-1 3240 1052 177 12 5.46296 3240 AAACCTGAGATAGTCA-1 1683 739 124 11 7.36780 1683 AAACCTGAGGCATGGT-1 2983 951 67 11 2.24606 2983 AAACCTGCAAGGTTCT-1 4181 1248 93 10 2.22435 4181 AAACCTGCAGGATTGG-1 2691 1350 183 10 6.80045 2691
Any help please.
What is the output of
ncol(sce)before running the clustering? My best guess is that your filtering removed all cells.Thanks, I realized I changed something upstream that affected this. Thanks again.