skills.rest
chenyhvvvv
chenyhvvvvCOMMUNITY
Explore all skills by this author →
GitHub

integration-bbknn

Modifies neighbor graph with BBKNN to batch-correct across multiple slices and outputs joint UMAP embedding and Leiden clusters.

3|Updated Apr 3, 2026
Search Tags:#batch-correction#umap#leiden#bbknn#slice-integration#joint-clustering
One-click install
npx skills add https://github.com/chenyhvvvv/STAT-agent --skill integration-bbknn
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: integration-bbknn
Source: https://github.com/chenyhvvvv/STAT-agent/tree/main/stat_agent/skills/integration-bbknn
Command: npx skills add https://github.com/chenyhvvvv/STAT-agent --skill integration-bbknn

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

BBKNN fixes batch effects across multiple slices by adjusting the neighbor graph, enabling joint clustering and visualization without altering the expression matrix.

Core Features & Use Cases

  • BBKNN-based batch correction across slices
  • Generates joint UMAP embedding and Leiden clusters
  • Lightweight and fast integration suitable for multi-slice studies

Quick Start

Load at least two slices in your session and run the BBKNN integration workflow to produce a batch-balanced neighbor graph and joint clustering.

Frequently Asked Questions about integration-bbknn

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

FAQPage Schema
How does BBKNN batch correction work for multi-slice integration?▼

BBKNN batch correction modifies the neighbor graph across multiple slices to balance batches, enabling joint clustering and visualization without altering the original expression matrix.

How do I generate a joint UMAP embedding and Leiden clustering across multiple slices?▼

You generate a joint UMAP embedding and Leiden clusters by applying BBKNN integration to combined multi-slice data, which outputs a batch-balanced neighbor graph in adata.obsp and joint embedding in adata.obsm.

Can I use BBKNN integration for joint clustering if I only have one slice loaded?▼

No, BBKNN integration requires at least two slices loaded in your session to perform batch correction and produce a batch-balanced neighbor graph for joint clustering.

What is the best way to perform slice integration without losing biological variation?▼

Using BBKNN to modify the neighbor graph is a lightweight method for slice integration that batch-corrects across slices while preserving biological variation in the expression matrix.

Does batch correction with BBKNN alter the expression matrix?▼

No, BBKNN batch correction does not alter the expression matrix; it modifies the neighbor graph to balance batches across slices, outputting the integrated graph in adata.obsp.

Related Skills

Explore More Data & Analytics Skills→
TianGzlabTianGzlab
155

spatial-integrate

Integrate multi-sample spatial transcriptomics data with Harmony, BBKNN, or Scanorama.

Official
Intermediate
chenyhvvvvchenyhvvvv
3

integration-scanorama

Correct batch effects across spatial slices using Scanorama integration.

Community
Advanced
ya-wayya-way

bio-single-cell-clustering

Cluster single-cell RNA-seq data with Seurat or Scanpy workflows.

Community
Advanced
HolobiomicsLabHolobiomicsLab
4

scvi-tools

Train scvi-tools models on AnnData for batch-corrected latent spaces and differential expression.

Official
Advanced