skills.rest
omicverse
omicverseOFFICIAL
Explore all skills by this author →
WebsiteGitHub

single-trajectory-analysis

Infer developmental trajectories and compute RNA velocity on single-cell AnnData objects.

32|5|Updated Mar 12, 2026
Search Tags:#trajectory#pseudotime#single-cell#scvelo#palantir#rna-velocity#latentvelo
One-click install
npx skills add https://github.com/omicverse/omicclaw --skill single-trajectory-analysis-omicverse
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: single-trajectory-analysis
Source: https://github.com/omicverse/omicclaw/tree/main/src/omicverse_skills/skills/single-trajectory
Command: npx skills add https://github.com/omicverse/omicclaw --skill single-trajectory-analysis-omicverse

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Provides a reproducible workflow to infer cellular lineages and directional dynamics from single-cell expression data, resolving branching trajectories and estimating pseudotime and fate probabilities without manual ad-hoc scripting.

Core Features & Use Cases

  • Unified trajectory toolbox: Guidance for PAGA, Palantir, and VIA to detect branches and terminal states.
  • RNA velocity integration: Instructions for scVelo, dynamo, latentvelo, and graphvelo backends and how to couple velocities to VIA for directionality refinement.
  • OmicVerse Velo API: Unified Velo class pipeline for filtering, moments, dynamics fitting, velocity calculation, graph construction, and embedding projection.
  • Downstream fate scoring and validation: Pseudotime overlays, fate probability export, metacell aggregation, CytoTRACE integration, and visualization best practices.
  • Use Case: Reconstruct T cell differentiation trajectories, validate roots with velocity arrows, and export pseudotime and fate probabilities for downstream marker discovery.

Quick Start

Run a unified Velo pipeline on a preprocessed AnnData to compute moments, fit dynamics, calculate velocities with your chosen backend, and project velocity embeddings to UMAP.

Frequently Asked Questions about single-trajectory-analysis

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

FAQPage Schema
How do I compute RNA velocity on a single-cell AnnData object?▼

To compute RNA velocity on a single-cell AnnData object, use the unified Velo API to run moments, fit dynamics, calculate velocities with your chosen backend, and project velocity embeddings to UMAP.

What is the best way to infer cellular lineages and branching trajectories from scRNA-seq data?▼

To infer cellular lineages and branching trajectories from scRNA-seq data, apply PAGA, Palantir, or VIA on a preprocessed AnnData to detect branches, identify terminal states, and estimate pseudotime.

Do I need spliced and unspliced layers in my AnnData to calculate RNA velocity?▼

Yes, you need spliced and unspliced layers in your AnnData object to calculate RNA velocity. Velocity backends including scVelo, dynamo, latentvelo, and graphvelo require these layers to infer directional dynamics.

Can I use latentvelo with GPU support for single-cell trajectory analysis?▼

Yes, you can use latentvelo with GPU support for single-cell trajectory analysis. Latentvelo requires torchdiffeq and optionally supports GPU acceleration to compute velocity on large-scale datasets.

Does this workflow support exporting fate probabilities and pseudotime for downstream marker discovery?▼

Yes, the workflow supports exporting fate probabilities and pseudotime for downstream marker discovery. You can overlay pseudotime, export fate probabilities, aggregate metacells, and integrate CytoTRACE for validation.

What should I do if my single-cell trajectory analysis requires coupling velocities for directionality refinement?▼

If your single-cell trajectory analysis requires coupling velocities for directionality refinement, couple the computed RNA velocities from your chosen backend to VIA, which refines directionality and improves branch probability estimation.

Related Skills

Explore More Data & Analytics Skills→
Estrella-231Estrella-231

scvelo

Infer RNA velocity and cell fate trajectories from single-cell RNA-seq AnnData.

Community
Advanced
chenyhvvvvchenyhvvvv
3

trajectory-pseudotime

Infer cell developmental trajectories and pseudotime ordering from gene expression data.

Community
Advanced
omicverseomicverse
13

omicverse-single-cell-rna-velocity

Analyze single-cell AnnData RNA velocity workflows and refine velocity outputs.

Official
Advanced
omicverseomicverse
13

omicverse-single-cell-trajectory-inference

Run OmicVerse single-cell trajectory inference on cluster-ready AnnData objects.

Official
Advanced