skills.rest
Starlitnightly
StarlitnightlyCOMMUNITY
Explore all skills by this author →
GitHub

BulkTrajBlend trajectory interpolation

Interpolate missing single-cell trajectory states using matched bulk RNA-seq data.

1.2k|145|Updated Mar 22, 2021
Search Tags:#bioinformatics#bulk RNA-seq#developmental biology#single-cell trajectory#beta-VAE#pseudotime#GNN#interpolation#omicverse
One-click install
npx skills add https://github.com/Starlitnightly/omicverse --skill bulktrajblend-trajectory-interpolation
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: BulkTrajBlend trajectory interpolation
Source: https://github.com/Starlitnightly/omicverse/tree/main/.claude/skills/bulk-trajblend-interpolation
Command: npx skills add https://github.com/Starlitnightly/omicverse --skill bulktrajblend-trajectory-interpolation

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires omicverse, scanpy, scvelo, anndata, matplotlib, numpy, pandas, torch, and includes references (resource) components.

What problem does it solves? Single-cell RNA-seq developmental trajectories often have gaps or missing intermediate cell states, limiting a complete understanding of cellular transitions. This Skill uses BulkTrajBlend to leverage bulk RNA-seq data to interpolate these missing states, providing a more continuous view of development.

Core Features & Use Cases

  • Synthetic Cell Generation: Deconvolute bulk RNA-seq samples to generate synthetic intermediate single cells.
  • GNN-Based Community Detection: Train Graph Neural Network (GNN) models to identify overlapping communities within the integrated data.
  • Trajectory Interpolation: Synthesize continuity and interpolate missing cell states to create smooth developmental trajectories.
  • Trajectory Analysis: Analyze and validate interpolated trajectories using VIA and PAGA for topological insights.
  • Use Case: Enhance a sparse single-cell trajectory of brain development by integrating matched bulk RNA-seq, generating intermediate cell types, and then re-analyzing the trajectory with VIA and PAGA to reveal finer transitions.

Quick Start

Use BulkTrajBlend to interpolate missing OPC states in my scRNA-seq trajectory using bulk RNA-seq data, then visualize the new trajectory with PAGA.

Frequently Asked Questions about BulkTrajBlend trajectory interpolation

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

FAQPage Schema
How do I interpolate missing states in single-cell developmental trajectories using bulk RNA-seq data?▼

BulkTrajBlend interpolates missing cell states by deconvoluting matched bulk RNA-seq samples into synthetic intermediate cells, then integrating them with single-cell data to bridge trajectory gaps and create continuous developmental progression.

Can I use bulk RNA-seq to fill gaps in my scRNA-seq trajectory analysis?▼

Yes. BulkTrajBlend leverages paired bulk and single-cell datasets to generate synthetic intermediate cell types via deconvolution, filling missing developmental stages that single-cell alone cannot capture.

What input data do I need to run trajectory interpolation with bulk and single-cell RNA-seq?▼

You need matched bulk RNA-seq counts (unscaled), single-cell data in AnnData format with a celltype_key annotation, and optionally target counts or TAPE-based estimates for beta-VAE configuration to generate synthetic cells.

How does BulkTrajBlend generate synthetic cells for trajectory analysis?▼

BulkTrajBlend deconvolutes bulk samples using matched single-cell references, then trains a beta-VAE on the deconvolved profiles to synthesize intermediate cell states that represent missing developmental transitions.

Can I validate interpolated trajectories after using bulk RNA-seq data?▼

Yes. After interpolation, you can analyze and validate the new trajectory using VIA and PAGA to reveal finer cell transitions and confirm topological consistency of the synthetic developmental states.

What dependencies do I need installed to use trajectory interpolation with bulk and single-cell data?▼

Core dependencies include omicverse, scanpy, scvelo, anndata, torch, numpy, pandas, and matplotlib for deconvolution, beta-VAE training, GNN community detection, and trajectory visualization.

Related Skills

Explore More Education & Research Skills→
omicverseomicverse
13

omicverse-bulk-celltype-deconvolution

Infer cell-type fractions from bulk RNA-seq using a paired single-cell reference.

Official
Advanced
omicverseomicverse
32

bulktrajblend-trajectory-interpolation

Generate intermediate single-cell profiles from bulk RNA-seq using beta-VAE and GNN community detection.

Official
Advanced
mkurmanmkurman
321

scvelo

Analyze RNA velocity in single-cell RNA-seq data with scVelo models.

Community
Advanced
TianGzlabTianGzlab
155

spatial-trajectory

Infer single-cell trajectories and pseudotime from AnnData or Loom files.

Official
Advanced