Challenges and Progress in RNA Velocity: Comparative Analysis Across Multiple Biological Contexts
Sarah Ancheta, Leah Dorman, Guillaume Le Treut, Abel Gurung, Greg Huber, Loïc A. Royer, Alejandro Granados, Merlin Lange
Abstract
Single-cell RNA sequencing is revolutionizing our understanding of cell state dynamics, allowing researchers to capture and quantify the transcriptomic profile of a single cell at a specific timepoint. Among the computational techniques used to predict cellular trajectories, RNA velocity has emerged as a predominant tool for modeling transcriptional dynamics. RNA velocity leverages the mRNA maturation process to generate velocity vectors that predict the likely future state of a cell, offering insights into cellular differentiation, aging, and disease progression.
Introduction
Single-cell RNA sequencing (scRNA-seq) has enabled the characterization of thousands of transcriptomic states, defined by distinct gene expression profiles across cells, and many computational methods have been developed to infer the state lineages. While some cell populations exist in equilibrium, others constantly change due to cell differentiation, environmental changes, cell cycle, or disease perturbations.
Materials and Methods:
Though local consistency with a method is an important metric in evaluating RNA velocity methods, alternatively, one can ask if the vector predictions from different velocity methods agree. We compare the cell-cell transition matrices from each method directly, in the shared cell-cell space of each dataset. Agreement between methods can help to identify lineages and cellular states with stronger velocity signals that correlate with biological relevance.
Discussion:
We evaluated the performance of five RNA velocity methods on three developmental datasets by analyzing their local consistency, method agreement, overlap of driver genes, and robustness to sequencing depth. Collectively, the RNA velocity methods identified known biological trajectories and important driver genes, with each method displaying varying levels of performance depending on the dataset and evaluation metric.
Acknowledgments:
We thank the Data Science and Royer teams of the Biohub for their helpful discussion. We are grateful for the review and feedback from Sandy Schmid, Jordão Bragantini, Joan Wong, and Yang-Joon Kim. We thank the Biohub and its donors, Priscilla Chan and Mark Zuckerberg, for funding this work. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Citation: Ancheta S, Dorman L, Le Treut G, Gurung A, Huber G, Royer LA, et al. (2026) Challenges and progress in RNA velocity: Comparative analysis across multiple biological contexts. PLoS Comput Biol 22(6): e1014303. https://doi.org/10.1371/journal.pcbi.1014303
Editor: Ilya Ioshikhes, CANADA
Received: May 13, 2025; Accepted: May 6, 2026; Published: June 1, 2026.
Copyright: © 2026 Ancheta et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant code is available here https://github.com/czbiohub-sf/comparison-RNAVelo. The data sets analyzed in this paper are from previously published research and are publicly available. Zebrafish raw sequencing data are available at NCBI’s SRA BioProject PRJNA940501. The raw dataset of pancreatic endocrinogenesis has been deposited under the accession number GSE132188.
Funding: Funding is provided by Biohub San Francisco, USA. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
