RegVelo: gene-regulatory-informed dynamics of single cells¶
RegVelo is an end-to-end framework to infer regulatory cellular dynamics through coupled splicing dynamics. See our RegVelo manuscript and documentation to learn more.
Feel free to open an issue if you encounter a bug, need help, have a suggestion, or notice any cases where regvelo doesn’t work as expected. Reports of unexpected behavior or edge cases are especially valuable and help us continue improving the package! We also maintain an FAQ that addresses some commonly encountered questions and unexpected behaviors. You are also welcome to contact the lead developer directly: weixu.wang@helmholtz-munich.de
RegVelo’s key applications¶
Estimate RNA velocity governed by gene regulation.
Infer latent time to indicate the cellular differentiation process.
Estimate intrinsic and extrinsic velocity uncertainty Gayoso et al. (2024).
Estimate regulon perturbation effects via CellRank framework (Lange et al. (2022), Weiler et al. (2024)).
Getting started¶
We have tutorials to help you get started.
Resources¶
RegVelo ships with several published datasets via regvelo.datasets, so you can try the pipeline without bringing your own data:
Zebrafish neural crest (Smart-seq3) —
rgv.datasets.zebrafish_nc()loads Smart-seq3 profiles across seven time points of neural crest development, paired with a prior GRN viargv.datasets.zebrafish_grn().Zebrafish neural crest (Perturb-seq) —
rgv.datasets.zebrafish_perturb()loads a single-cell CRISPR screening dataset (12,393 cells) with TF knockouts, useful for benchmarking in silico perturbation predictions against ground truth.Human embryonic hindbrain —
rgv.datasets.hindbrain()loads single-cell RNA-seq data from the developing human hindbrain, paired with a prior GRN viargv.datasets.hindbrain_grn().
See the datasets API reference for the full list, including the murine neural crest, human limb, and Schwann cell datasets.
Installation¶
You need to have Python 3.10 or newer installed on your system.
There are several options to intall regvelo:
Install the latest release of
regvelofrom PyPI via
pip install regvelo
Install the latest development version via
pip install git+https://github.com/theislab/regvelo.git@main
RegVelo agentic workflow via Biomni lab¶
RegVelo pipeline can be run using an agentic workflow via Biomni lab, which allows user to upload custom datasets and run full analysis using a chat-based approach.
Citation¶
If you find RegVelo useful for your research, please consider citing our work as:
@article{wang2026regvelo,
title={RegVelo: gene-regulatory-informed dynamics of single cells},
author={Wang, Weixu and Hu, Zhiyuan and Weiler, Philipp and Mayes, Sarah and Lange, Marius and Wang, Jingye and Xue, Zhengyuan and Sauka-Spengler, Tatjana and Theis, Fabian J},
journal={Cell},
year={2026},
publisher={Elsevier},
doi={10.1016/j.cell.2026.04.022}
}