How does scvi-tools differ from Scanpy?

Scanpy is primarily used for standard analysis pipelines and data visualization, while scvi-tools focuses on deep generative probabilistic models for complex tasks like integration and deconvolution.

What is scvi-tools best used for?

It is best for advanced modeling of single-cell data, specifically for batch effect correction, multi-modal integration, and differential expression with uncertainty.

Does this skill require a GPU?

While not strictly required, scvi-tools is optimized for GPU acceleration via PyTorch, which is highly recommended for processing large single-cell datasets efficiently.

Can I use this for spatial transcriptomics?

Yes, this skill includes support for spatial models like DestVI and Tangram for cell type deconvolution and mapping single-cell data to spatial locations.

Does it support raw count data?

Yes, scvi-tools is specifically designed to work with raw, unnormalized count data to maintain statistical integrity during probabilistic modeling.

scvi-tools Single-Cell Analysis

Name: scvi-tools Single-Cell Analysis
Author: Yezez9

byYezez9

0•

数据科学与机器学习

Analyzes single-cell omics data using deep generative models and probabilistic inference for batch correction and multimodal integration.

scvi-tools provides a comprehensive Python framework for applying deep generative models to single-cell genomics, enabling researchers to perform advanced probabilistic modeling of RNA-seq, ATAC-seq, and spatial transcriptomics data. Built on PyTorch and PyTorch Lightning, it automates complex tasks like batch correction, differential expression with uncertainty quantification, and multimodal data integration, making it an essential tool for high-dimensional biological data analysis within the scanpy ecosystem.

主要功能

01Probabilistic batch correction and data integration

02Multimodal analysis for CITE-seq and Multiome data

03Deep generative modeling for scRNA-seq and scATAC-seq

04Semi-supervised cell type annotation and transfer learning

05Differential expression with built-in uncertainty quantification

060 GitHub stars

使用场景

01Mapping single-cell profiles to spatial transcriptomics data for cell-type deconvolution

02Jointly analyzing protein and RNA expression in CITE-seq data using totalVI

03Integrating multiple single-cell datasets across different batches or experiments

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add yezez9/research-agent scvi-tools

For use in Claude.ai and ChatGPT