In bioinfocz/scdrake: A pipeline for droplet-based single-cell RNA-seq data secondary analysis implemented in the drake Make-like toolkit for R language

Detecting empty droplets

DropletUtils::emptyDrops() computes Monte Carlo p-values based on a Dirichlet-multinomial model of sampling molecules into droplets. emptyDrops() assumes that libraries with total UMI counts below a certain threshold (r cfg$EMPTY_DROPLETS_LOWER by default) correspond to empty droplets. These are used to estimate the ambient expression profile against which the remaining libraries are tested. Under this definition, these low-count libraries cannot be cell-containing droplets and are excluded from the hypothesis testing.

Number of non-empty cells: r sce_valid_cells_info$dim[2]

Droplets detected as cells should show up with large negative log-probabilities or very large total UMI counts (based on the knee point).

is_cell <- empty_droplets$FDR <= cfg$EMPTY_DROPLETS_FDR_THRESHOLD
plot(empty_droplets$Total, -empty_droplets$LogProb, col = ifelse(is_cell, "red", "black"), xlab = "Total UMI count", ylab = "-Log Probability")

Droplets with empty-droplet FDR > r cfg$EMPTY_DROPLETS_FDR_THRESHOLD have been removed. Filtered dataset summary:

cat(sce_valid_cells_info$str)

r scdrake::format_used_functions("DropletUtils::emptyDrops()")

bioinfocz/scdrake documentation built on Sept. 19, 2024, 4:43 p.m.