Preprocessing choices affect RNA velocity results for droplet scRNA-seq data

Author summary Applied to single-cell RNA-seq data, RNA velocity analysis provides a way to estimate the rate of change of the gene expression levels in individual cells. This, in turn, enables estimation of what the gene expression profile of each cell will look like a short time into the future and lets researchers infer likely developmental relationships among different types of cells in a tissue. An important first step in this type of analysis consists of estimating the expression levels of unspliced pre-mRNA as well as mature mRNA in each cell. Several methods are available for this purpose, and in this study we perform a comparison of these tools and highlight respective advantages and disadvantages. We envision that the results will be informative for researchers performing RNA velocity analysis, as well as to guide future developments in the evaluated and newly proposed quantification methods. Experimental single-cell approaches are becoming widely used for many purposes, including investigation of the dynamic behaviour of developing biological systems. Consequently, a large number of computational methods for extracting dynamic information from such data have been developed. One example is RNA velocity analysis, in which spliced and unspliced RNA abundances are jointly modeled in order to infer a 'direction of change' and thereby a future state for each cell in the gene expression space. Naturally, the accuracy and interpretability of the inferred RNA velocities depend crucially on the correctness of the estimated abundances. Here, we systematically compare five widely used quantification tools, in total yielding thirteen different quantification approaches, in terms of their estimates of spliced and unspliced RNA abundances in five experimental droplet scRNA-seq data sets. We show that there are substantial differences between the quantifications obtained from different tools, and identify typical genes for which such discrepancies are observed. We further show that these abundance differences propagate to the downstream analysis, and can have a large effect on estimated velocities as well as the biological interpretation. Our results highlight that abundance quantification is a crucial aspect of the RNA velocity analysis workflow, and that both the definition of the genomic features of interest and the quantification algorithm itself require careful consideration.

Automated summary

RNA velocity analysis estimates the rate of change of gene expression levels in single cells, enabling prediction of future gene expression profiles and developmental relationships. This study compares five widely used quantification tools in five experimental datasets, highlighting differences in estimates and their impact on downstream analysis, emphasizing the importance of careful consideration of genomic features and quantification algorithms in RNA velocity analysis workflow.