Single-Cell Sequencing Reveals Lineage-Specific Dynamic Genetic Regulation of Gene Expression During Human Cardiomyocyte Differentiation

The impact of genetic regulatory elements on gene expression can vary across cell states. During dynamic processes such as cellular differentiation, cells transition through multiple cell states, and may differentiate toward multiple terminal cell types. We collected time-series single-cell RNA-sequencing data from 19 human cell lines, capturing 7 time points along a 16-day differentiation from induced pluripotent stem cells to cardiomyocytes. We used unsupervised clustering, marker gene expression patterns, and pseudotime inference methods to map individual cells to a position along one of two bifurcating differentiation trajectories that were inferred from the data. We then identified genetic effects on gene regulation with varying effects across these trajectories. We identified hundreds of dynamic eQTLs that change significantly across pseudotime, including many variants whose effects are specific to one of the two lineages. We then re-analyzed previously collected bulk data, and used cell state information to infer cell type interaction eQTLs in bulk, assigning previously identified dynamic eQTLs to one of the newly characterized cellular trajectories. Overall design: We obtained droplet-based single-cell RNA-seq data for 19 Yoruba individuals at 7 time points each during the differentiation from iPSC to cardiomyocyte. There are a total of 57 collections, each containing cells pooled from 3 human cell lines.