kwp-bio-research-scvi-tools

Execute deep learning-based single-cell analysis workflows with scvi-tools and scanpy.

7|5|Updated May 7, 2026

Search Tags:#deep learning #bioinformatics #genomics #single-cell #batch correction #scvi-tools

Ready to run

npx skills add https://github.com/14790897/MiQi --skill kwp-bio-research-scvi-tools

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: kwp-bio-research-scvi-tools
Source: https://github.com/14790897/MiQi/tree/main/miqi/skills/kwp/bio-research/scvi-tools
Command: npx skills add https://github.com/14790897/MiQi --skill kwp-bio-research-scvi-tools

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires scvi-tools, scanpy, anndata, numpy, pandas, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This skill addresses the complexity of analyzing high-dimensional single-cell data by providing a standardized interface for probabilistic deep learning models, removing the need for manual implementation of complex genomic workflows.

Core Features & Use Cases

Multi-modal Integration: Seamlessly process scRNA-seq, ATAC-seq, CITE-seq, and spatial transcriptomics data using specialized variational autoencoders.
Batch Correction & Mapping: Perform robust data integration and reference mapping to ensure biological signals are preserved across different experimental batches or technologies.
Use Case: Researchers can use this skill to integrate multiple scRNA-seq datasets, perform label transfer from a reference atlas, or deconvolve spatial transcriptomics data with minimal boilerplate code.

Quick Start

Use the scvi-tools skill to train an scVI model on the provided single-cell dataset to perform batch correction and latent space embedding.

Frequently Asked Questions about kwp-bio-research-scvi-tools

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

FAQPage Schema

How do I perform batch correction on scRNA-seq data using deep learning?▼

Batch correction on scRNA-seq data is executed by training variational autoencoders to produce latent space embeddings that align experimental batches. This deep learning approach removes technical variation while preserving biological signals for accurate downstream analysis.

Can I integrate ATAC-seq and CITE-seq data for multi-modal single-cell analysis?▼

Yes, you can integrate ATAC-seq and CITE-seq data for multi-modal single-cell analysis using specialized variational autoencoders. The framework processes chromatin accessibility and protein denoising count matrices simultaneously to model complex cellular states across different genomic technologies.

Do I need scanpy and anndata installed to run scvi-tools workflows?▼

Yes, you need scanpy and anndata installed because the workflows require these environments to manage high-dimensional count matrices. These dependencies provide the essential data structures and preprocessing functions required to execute variational inference on single-cell genomic data.

What is the best way to deconvolve spatial transcriptomics data with single-cell references?▼

The best way to deconvolve spatial transcriptomics data is by using deep learning-based reference mapping to transfer labels from a single-cell atlas. This probabilistic approach estimates cell type proportions within spatial spots by leveraging variational inference on high-dimensional counts.

How does variational inference handle high-dimensional single-cell count matrices?▼

Variational inference handles high-dimensional single-cell count matrices by training deep learning models to compress data into a structured latent space. This probabilistic approach effectively models complex distributions, enabling robust data integration and denoising for genomic workflows.