scvelo

Infer RNA velocity from single-cell RNA-seq using spliced and unspliced mRNA kinetics.

4|Updated Mar 2, 2026
One-click install
npx skills add https://github.com/shushuzn/Rairos --skill scvelo-shushuzn
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: scvelo
Source: https://github.com/shushuzn/Rairos/tree/main/skills/scvelo
Command: npx skills add https://github.com/shushuzn/Rairos --skill scvelo-shushuzn

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires scvelo, scanpy, numpy, matplotlib, and includes scripts (resource) and references (resource) components.

What problem does it solve?

scVelo helps you infer RNA velocity from snapshot single-cell RNA-seq data, revealing the direction of cell state transitions and estimating latent timing without requiring time-course experiments.

Core Features & Use Cases

  • RNA velocity from spliced/unspliced dynamics: Estimate whether genes are upregulated or downregulated per cell by modeling unspliced versus spliced mRNA kinetics.
  • Trajectory direction and latent time: Reconstruct developmental trajectories and compute shared latent time (pseudotime-like ordering) from the dynamical model.
  • Driver gene identification: Rank genes whose kinetic dynamics best explain observed transitions, and quantify model confidence for coherence.

Quick Start

Use scvelo to compute RNA velocity and stream arrows on your existing UMAP by running the dynamical workflow on an AnnData object that contains layers["spliced"] and layers["unspliced"].

Frequently Asked Questions about scvelo

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I infer RNA velocity from single-cell RNA-seq data without time-course experiments?

RNA velocity is inferred from snapshot single-cell RNA-seq data by modeling unspliced and spliced mRNA kinetics to estimate gene-wise state transition directions without requiring time-course experiments.

What is needed to compute trajectory direction and latent time in scvelo?

Computing trajectory direction and latent time requires an AnnData object containing spliced and unspliced layers, applying the dynamical model to reconstruct developmental trajectories and estimate shared latent time.

Can I use scvelo with my existing UMAP and scanpy workflow?

Yes, scvelo integrates with scanpy workflows by computing RNA velocity and projecting stream arrows onto your existing UMAP visualization using an AnnData object.

What is the difference between stochastic and dynamical models for RNA velocity?

The stochastic model provides fast RNA velocity estimation, while the dynamical model is recommended for accurately computing latent time, trajectory directionality, and driver gene ranking.

How do I identify driver genes across differentiation processes using scvelo?

Driver genes are identified by ranking genes whose kinetic dynamics best explain observed cell state transitions and quantifying model confidence for coherence across developmental or differentiation processes.