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

single-cell-multi-omics-integration

Integrate paired and unpaired single-cell AnnData modalities into harmonized embeddings and joint latent factors.

32|5|Updated Mar 12, 2026
Search Tags:#integration#single-cell#glue#multi-omics#simba#mofa#tosica
One-click install
npx skills add https://github.com/omicverse/omicclaw --skill single-cell-multi-omics-integration-omicverse
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: single-cell-multi-omics-integration
Source: https://github.com/omicverse/omicclaw/tree/main/src/omicverse_skills/skills/single-multiomics
Command: npx skills add https://github.com/omicverse/omicclaw --skill single-cell-multi-omics-integration-omicverse

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill removes the complexity of combining scRNA, scATAC, and other single-cell modalities into harmonized embeddings or joint latent factors so researchers can perform integrated downstream analyses without manually reconciling modalities, batches, or annotation formats.

Core Features & Use Cases

  • Paired multi-omics factorization (MOFA): Learn shared and modality-specific factors from matched cell measurements to interpret cross-modality variance.
  • Unpaired alignment then factorization (GLUE → MOFA): Align disparate RNA and ATAC experiments before downstream joint factor analysis.
  • Batch integration for single modalities (SIMBA): Build graph embeddings and apply batch correction across studies.
  • Reference label transfer (TOSICA): Transfer cell-type labels from annotated references with pathway-masked interpretability.
  • Trajectory cartography (StaVIA/VIA): Infer pseudotime and lineage structure using preprocessed data and velocity information.
  • Use Case: Combine three pancreas cohorts with mixed modalities, correct batch effects, transfer annotations from a curated reference, and compute trajectories for developmental inference.

Quick Start

Integrate my paired scRNA and scATAC AnnData files using MOFA to produce joint factors and return variance explained plots.

Frequently Asked Questions about single-cell-multi-omics-integration

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

FAQPage Schema
How do I integrate paired scRNA and scATAC single-cell datasets for joint analysis?▼

Paired scRNA and scATAC integration uses MOFA to learn shared and modality-specific factors from matched cell measurements. This produces harmonized joint latent factors that enable cross-modality variance interpretation and integrated downstream analysis.

What is the best way to align unpaired scRNA and scATAC experiments before factorization?▼

Unpaired scRNA and scATAC alignment uses GLUE to integrate disparate experiments before downstream joint factor analysis. GLUE produces harmonized embeddings that reconcile separate RNA and ATAC modalities into a shared latent space for subsequent MOFA factorization.

Can I transfer cell-type labels from an annotated reference to my single-cell dataset?▼

Reference label transfer uses TOSICA to transfer cell-type labels from annotated references with pathway-masked interpretability. This requires preprocessed AnnData inputs with completed PCA and neighbor graphs to accurately map labels onto query cells.

Do I need to preprocess my AnnData files before running multi-omics integration?▼

Multi-omics integration requires per-modality AnnData inputs with upstream preprocessing completed. You must perform HVG selection, PCA, and neighbor graph construction beforehand, as integration methods like MOFA, GLUE, and SIMBA operate on these precomputed representations.

How does single-cell batch correction work across multiple studies with mixed modalities?▼

Single-cell batch correction across studies uses SIMBA to build graph embeddings that reconcile batch effects and mixed modalities. This produces harmonized embeddings that remove study-specific technical variation while preserving biological signal across integrated cohorts.

When should I use VIA trajectory inference in my single-cell multi-omics workflow?▼

VIA trajectory inference applies after integration to infer pseudotime and lineage structure using preprocessed data and velocity information. Use StaVIA when you need trajectory cartography for developmental inference across harmonized multi-omics embeddings.

Related Skills

Explore More Education & Research Skills→
omicverseomicverse
13

omicverse-single-cell-preprocessing

Convert OmicVerse single-cell AnnData objects into cluster-ready preprocessing and marker-discovery workflows.

Official
Advanced
zongtingweizongtingwei
25

trajectory-lineage

Infer pseudotime, lineage branches, and state transitions from single-cell data using Scanpy.

Community
Intermediate
omicverseomicverse
1.2k

single-cell-multiomics-integration

Integrate paired and unpaired single-cell multi-omics datasets using MOFA, GLUE, SIMBA, TOSICA, and StaVIA.

Official
Advanced
zongtingweizongtingwei
25

multiome-scatac

Integrate single-cell RNA and ATAC data into unified multimodal embeddings.

Community
Advanced