Variable RNA sampling biases mediate concordance of single-cell and nucleus sequencing across cell types
A mechanistic understanding of the biological and technical factors that impact cell and nuclear transcript measurements is essential to designing, analyzing, and interpreting single-cell and single-nucleus RNA sequencing experiments. RNA sampling in nuclei and cells is fundamentally different as nuclei contain the same pre-mRNA population as cells, yet contain a small subset of the largely-cytoplasmic mRNAs. Nonetheless, early studies argued that including pre-mRNA in single-nucleus analysis led to results comparable to cellular samples. However, typical bioinformatic workflows do not distinguish between pre-mRNA and mRNA when analyzing gene expression, and variation in the relative abundance of pre-mRNA and mRNA across cell types has received limited attention. These gaps are especially important given that incorporating pre-mRNA in routine gene expression analysis is now commonplace for both assays, despite known gene length bias in pre-mRNA capture. Here, we reanalyze public datasets from mouse and human to describe the mechanisms and contrasting effects of mRNA and pre-mRNA sampling in single-cell and nucleus RNA-seq. We disentangle the roles of bioinformatic processing, assay choice, and biological variability on measured gene expression and marker gene selection. We show that pre-mRNA levels vary considerably among cell types, which mediates the degree of gene length bias within and between assays and limits the generalizability of a recently-published normalization method intended to correct for this bias. As an alternative solution, we demonstrate the applicability of an existing post hoc gene length-based correction method developed for conventional RNA-seq gene set enrichment analysis. Finally, we show that the inclusion of pre-mRNA in bioinformatic processing can impart a larger effect on gene expression estimates than the choice of cell versus nuclear assay, which is pivotal to the effective reuse of existing data. Broadly, these analyses advance our understanding of the biological and technical factors underlying variation in single-cell and single-nucleus RNA-seq experiments to promote more informed choices in experimental design, data analysis, and data sharing and reuse.