Efficient and accurate mixed model association tool for single-cell eQTL analysis.

SAIGE-QTL, a Poisson mixed model for single-cell read counts, identifies 48.8% more eGenes than pseudobulk analysis in 1.2 million immune cells.

• Why it matters: Single-cell eQTL mapping is challenged by cell-type–specific genetic effects, read-count–appropriate modeling, and scaling while handling donor and cell covariates.
• What they did: The method fits a Poisson mixed model that accounts for donor-level and cell-level covariates and unequal cell numbers per donor, and was evaluated in simulations and three single-cell datasets.
• The result: SAIGE-QTL improves statistical efficiency and eGene discovery versus pseudobulk, supports genome-wide trans-eQTL mapping, and enables rare variant set-based testing with associations independent of common eQTL signals.

AI summary of the abstract · may contain errors · not medical advice · read the paper.