What data format is required for these models?

The skill follows the scvi-tools standard of using AnnData objects, specifically requiring raw, unnormalized count data for optimal model performance.

Is GPU acceleration supported in these workflows?

Yes, the skill includes patterns for utilizing PyTorch and PyTorch Lightning to enable GPU acceleration for large-scale single-cell datasets.

Can I use this for batch correction and integration?

Absolutely. scvi-tools is designed to handle technical variation through covariate registration and latent representation learning using models like scVI.

Does it support multimodal data integration?

Yes, it provides guidance for models like totalVI (CITE-seq) and MultiVI (paired/unpaired multi-omics) to jointly analyze multiple data types.

Does this skill support spatial transcriptomics?

Yes, it includes specialized models like DestVI and Stereoscope for spatial deconvolution, mapping, and cell-environment relationship analysis.

scvi-tools for Single-Cell Analysis

Name: scvi-tools for Single-Cell Analysis
Author: BbgnsurfTech

byBbgnsurfTech

•

Data Science & ML

Implements probabilistic deep learning models for comprehensive single-cell omics data analysis and multimodal integration.

About

This skill empowers Claude to perform advanced single-cell genomics workflows using the scvi-tools framework, covering tasks from batch correction and dimensionality reduction to spatial transcriptomics and multimodal data integration. It provides expert guidance on implementing deep generative models like scVI, scANVI, and totalVI, ensuring rigorous probabilistic modeling and uncertainty quantification for complex biological datasets within the Python data science ecosystem.

Key Features

Spatial transcriptomics deconvolution and mapping models
3 GitHub stars
Multimodal and multi-omic data integration workflows
Probabilistic modeling for scRNA-seq, scATAC-seq, and CITE-seq
Automated batch correction and latent representation extraction
Seamless integration with AnnData and Scanpy ecosystems

Use Cases

Integrating multiple single-cell datasets while removing technical batch effects
Performing differential expression analysis using Bayesian hypothesis testing
Mapping cell types in spatial transcriptomics data through probabilistic deconvolution

About

Key Features

Spatial transcriptomics deconvolution and mapping models
3 GitHub stars
Multimodal and multi-omic data integration workflows
Probabilistic modeling for scRNA-seq, scATAC-seq, and CITE-seq
Automated batch correction and latent representation extraction
Seamless integration with AnnData and Scanpy ecosystems

Use Cases

Integrating multiple single-cell datasets while removing technical batch effects
Performing differential expression analysis using Bayesian hypothesis testing
Mapping cell types in spatial transcriptomics data through probabilistic deconvolution