Estimate Genetic Sex from Expression Counts

Infers the genetic sex of samples, donors or cells from bulk or single-cell RNA sequencing counts. Accepts dense or sparse matrices, delimited text files, 10x Genomics outputs, and 'Seurat', 'SingleCellExperiment' and 'SummarizedExperiment' objects.

SexSeek

SexSeek estimates genetic sex from bulk or single-cell expression counts, using sex-chromosome marker expression and X-inactivation markers. Each sample, donor or cell is called male, female, uncertain, possible_mixed or unknown, with confidence metrics.

Install

install.packages("SexSeek")

# development version
remotes::install_github("saketlab/SexSeek")

Use

library(SexSeek)

path <- system.file("extdata", "example_counts.tsv.gz", package = "SexSeek")
counts <- as.matrix(read.delim(path, row.names = 1))
EstimateSex(counts, group = colnames(counts))

EstimateSex(counts)                          # matrix, sparse matrix, data frame
EstimateSex("counts.tsv.gz")                 # delimited file
EstimateSex("filtered_feature_bc_matrix/")   # 10x MatrixMarket directory
EstimateSex("sample.h5")                     # 10x HDF5
EstimateSex(seurat_obj, group = "donor")     # aggregate by a metadata column

Species is detected from the gene identifiers; pass species = "mouse" (common or scientific name) to set it.

Genevintage annotations

You can provide your own set of sex genes or fetch it from genevintage:

install.packages("genevintage")
sex_genes <- SexChromosomeGenes("human", release = 116, source = "ensembl")
EstimateSex(counts, species = "human", annotation = sex_genes)
EstimateSex(counts, species = "human", annotation = "genevintage")