Expression quantitative trait loci (eQTL) mapping is the search for genomic loci associated with gene expression. Traditional methods perform eQTL mapping on bulk tissue gene expression without accounting for the underlying sources of gene expression per cell type. Attempts to perform ``cell type-aware'' eQTL mapping commonly use linear models that transform the gene expression while adjusting for the genotype, a proposed measure for cell type prevalence (e.g. a marker gene's expression), and their interaction and proceed to perform hypothesis testing.
Our R package performs cell type-specific eQTL mapping on bulk RNA-sequencing samples by jointly modeling total read count (TReC) and allele-specific read count (ASReC). Our method, CSeQTL, introduces novel features and fully extends TReCASE [HTML, PDF, Software] and pTReCASE [HTML, PDF, Software] methodologies.
Example of total read counts and allele-specific or haplotype read counts.
- TReC data
- ASReC data (haplotype 1 and 2 read counts)
- Phased genotypes
- Sample covariates:
- observed confounders,
- genotype principal components (PCs),
- latent batch effects (derived from residual TReC PCs)
- Cell type proportions
Click to expand!
req_packs = c("devtools","Rcpp","RcppArmadillo",
"smarter","CSeQTL")
build_vign = ifelse(Sys.getenv("RSTUDIO_PANDOC") == "",FALSE,TRUE)
for(pack in req_packs){
chk_pack = tryCatch(find.package(pack),
error = function(ee){NULL})
if( !is.null(chk_pack) ){
library(package = pack,
character.only = TRUE)
next
}
if( pack %in% c("smarter","CSeQTL") ){
repo = sprintf("pllittle/%s",pack)
install_github(repo = repo,
build_vignettes = build_vign,
dependencies = TRUE)
} else {
install.packages(pkgs = pack,
dependencies = TRUE)
}
}
# An Introduction
vignette(topic = "intro",package = "CSeQTL")Little, P., Liu, S., Zhabotynsky, V., Li, Y., Lin, D.Y., Sun, W. (2023). A computational method for cell type-specific expression quantitative trait loci mapping using bulk RNA-seq data. Nature Communications. [HTML, SUPP]