A robust model for cell type-specific interindividual variation in single-cell RNA sequencing data.

Area of Science:

• Genomics
• Computational Biology
• Developmental Biology

Background:

• Single-cell RNA sequencing (scRNA-seq) typically analyzes average gene expression per cell type.
• Inter-individual variation within cell types is often overlooked in scRNA-seq studies.
• Understanding cell type-specific variation is crucial for complex trait genetics and cell biology.

Purpose of the Study:

• Develop a novel statistical model, CTMM (Cell Type-specific linear Mixed Model), to detect and quantify inter-individual variation in scRNA-seq data.
• Assess the model's performance and statistical power using simulations.
• Apply the model to human induced pluripotent stem cell differentiation data.

Main Methods:

• Developed the Cell Type-specific linear Mixed Model (CTMM) for analyzing scRNA-seq data.
• Conducted extensive simulations to validate CTMM's accuracy and power.
• Applied CTMM to human iPSC differentiation data to analyze transcriptomic variation across donors and differentiation stages.

Main Results:

• CTMM demonstrated powerful and unbiased detection of cell type-specific inter-individual variation in simulations.
• Analysis of human iPSC differentiation revealed that nearly all donor-specific transcriptomic variability is linked to differentiation stage.
• Identified 85 genes with significant stage-specific variation not apparent in mean expression, and partitioned interindividual covariance to model differentiation trajectories.

Conclusions:

• CTMM is an effective tool for characterizing cell type-specific variation in scRNA-seq data.
• The model highlights the critical role of differentiation stage in shaping inter-individual transcriptomic differences.
• CTMM provides novel insights into cell type-specific biology and genetic underpinnings of complex traits.