Does this skill require raw count data for analysis?

Yes, scvi-tools models are specifically designed to work with unnormalized raw count data to accurately model technical variation and sequencing depth.

What modalities are supported by this skill?

It supports a wide range of modalities including scRNA-seq, scATAC-seq, CITE-seq, multi-omics, methylation, and flow/mass cytometry.

Is GPU acceleration supported for large datasets?

Yes, the skill provides guidance on utilizing GPU acceleration via PyTorch and PyTorch Lightning to efficiently process millions of cells.

Does this skill integrate with the Scanpy ecosystem?

Yes, the workflow is entirely AnnData-centric, allowing for seamless transitions between scvi-tools probabilistic modeling and Scanpy visualization or clustering.

Can it handle spatial transcriptomics datasets?

Absolutely. This skill includes support for models like DestVI, Stereoscope, and Tangram for spatial deconvolution, mapping, and cell-environment analysis.

scvi-tools Single-Cell Analysis

Name: scvi-tools Single-Cell Analysis
Author: jimmc414

byjimmc414

•

324

Data Science & ML

Analyzes single-cell omics data using deep generative models for batch correction, multi-omic integration, and probabilistic modeling.

About

scvi-tools is a comprehensive Python framework designed for the probabilistic modeling of single-cell genomics data. It leverages PyTorch-based deep generative models and variational inference to perform complex tasks like batch correction, dimensionality reduction, and multi-omic integration across modalities such as scRNA-seq, scATAC-seq, and spatial transcriptomics. This skill enables Claude to guide researchers through end-to-end bioinformatics workflows, from initial data setup in AnnData objects to advanced differential expression analysis, cell type annotation, and spatial deconvolution, ensuring best practices in statistical genomics are followed.

Key Features

Multimodal data integration for paired and unpaired datasets
Scalable batch correction and data integration using scVI and scANVI
324 GitHub stars
Probabilistic modeling for scRNA-seq, scATAC-seq, and CITE-seq data
Advanced differential expression analysis based on generative model posteriors
Spatial transcriptomics deconvolution and mapping with DestVI

Use Cases

Analyzing spatially-resolved transcriptomics to map cell-environment relationships and deconvolve spots
Performing automated cell type annotation and transfer learning on large-scale atlas data
Integrating heterogeneous single-cell datasets across different experiments or technologies

About

Key Features

Multimodal data integration for paired and unpaired datasets
Scalable batch correction and data integration using scVI and scANVI
324 GitHub stars
Probabilistic modeling for scRNA-seq, scATAC-seq, and CITE-seq data
Advanced differential expression analysis based on generative model posteriors
Spatial transcriptomics deconvolution and mapping with DestVI

Use Cases

Analyzing spatially-resolved transcriptomics to map cell-environment relationships and deconvolve spots
Performing automated cell type annotation and transfer learning on large-scale atlas data
Integrating heterogeneous single-cell datasets across different experiments or technologies