Fast calculation of scSE score

Calculates Single-Cell Signature Explorer (Pont et al., 2019) scores using plaid back-end. The computation is 10-100x faster than the original code.

Usage

replaid.scse(
  X,
  matG,
  removeLog2 = NULL,
  scoreMean = FALSE,
  assay = "logcounts",
  min.genes = 5,
  max.genes = 500
)

Arguments

X

Gene or protein expression matrix. Generally log transformed. See details. Genes on rows, samples on columns. Also accepts SummarizedExperiment or SingleCellExperiment objects.

matG

Gene sets sparse matrix. Genes on rows, gene sets on columns. Also accepts BiocSet objects or GMT lists.

removeLog2

Logical for whether to remove the Log2, i.e. will apply power transform (base2) on input (default TRUE).

scoreMean

Logical for whether computing sum or mean as score (default FALSE).

assay

Character: assay name for Bioconductor objects. Default "logcounts".

min.genes

Integer: minimum genes per gene set. Default 5.

max.genes

Integer: maximum genes per gene set. Default 500.

Value

Matrix of single-sample scSE enrichment scores. Gene sets on rows, samples on columns.

Details

Computing the scSE requires running plaid on the linear (not logarithmic) score and perform additional normalization by the total UMI per sample. We have wrapped this in a single convenience function:

To replicate the original "sum-of-UMI" scSE score, set removeLog2=TRUE and scoreMean=FALSE. scSE and plaid scores become more similar for removeLog2=FALSE and scoreMean=TRUE.

We have extensively compared the results from replaid.scse and from the original scSE (implemented in GO lang) and we showed almost identical results in the score, logFC and p-values.

Examples

# Create example expression matrix (log-transformed)
set.seed(123)
X <- log2(matrix(rpois(500, lambda = 10) + 1, nrow = 50, ncol = 10))
rownames(X) <- paste0("GENE", 1:50)
colnames(X) <- paste0("Sample", 1:10)

# Create example gene sets
gmt <- list(
  "Pathway1" = paste0("GENE", 1:15),
  "Pathway2" = paste0("GENE", 10:25)
)
matG <- gmt2mat(gmt)

# Compute scSE scores (original method)
scores <- replaid.scse(X, matG, removeLog2 = TRUE, scoreMean = FALSE)
#> [replaid.scse] Converting data to linear scale (removing log2)...
print(scores[1:2, 1:5])
#>           Sample1  Sample2  Sample3  Sample4  Sample5
#> Pathway2 28.82012 30.90909 36.29764 32.15613 31.62879
#> Pathway1 32.68859 27.81818 31.39746 31.22677 31.81818

# Compute scSE scores (mean method)
scores_mean <- replaid.scse(X, matG, removeLog2 = TRUE, scoreMean = TRUE)
#> [replaid.scse] Converting data to linear scale (removing log2)...
print(scores_mean[1:2, 1:5])
#>            Sample1   Sample2   Sample3   Sample4   Sample5
#> Pathway2 0.7584241 0.8133971 0.9552011 0.8462140 0.8323365
#> Pathway1 0.9080163 0.7727273 0.8721516 0.8674102 0.8838384