scDesignPop generates realistic population-scale single-cell RNA-seq for power analysis, benchmarking, and privacy...

Area of Science:

• Genomics
• Computational Biology
• Bioinformatics

Background:

• Single-cell RNA sequencing (scRNA-seq) with genotyping identifies cell-type-specific genetic associations (eQTLs).
• Challenges include high costs, lack of consensus in analysis methods, and privacy risks with data sharing.

Purpose of the Study:

• Introduce scDesignPop, a simulator for realistic population-scale scRNA-seq data with genetic effects.
• Address challenges in cost, analysis standardization, and privacy for eQTL studies.

Main Methods:

• scDesignPop models cell- and individual-level covariates and cell-type-specific eQTLs (cts-eQTLs).
• It supports real or synthetic genotypes and was validated using OneK1K and CLUES cohorts.
• Comparison with splatPop on 4 qualitative and 16 quantitative metrics.

Main Results:

• scDesignPop better preserves eQTL effects and gene-gene dependencies within cell types compared to splatPop.
• The simulator closely recapitulates characteristics of reference scRNA-seq data.
• Validated across multiple metrics, demonstrating its effectiveness.

Conclusions:

• scDesignPop facilitates power analysis for experimental design in cell types.
• Enables benchmarking of single-cell eQTL mapping methods using user-defined ground truths.
• Mitigates re-identification risks via synthetic data generation while preserving cts-eQTL effects.