What data format does scvi-tools require?

The framework primarily uses the AnnData format and requires raw, unnormalized count data for its probabilistic models to perform correctly.

Does this skill support GPU acceleration?

Absolutely. scvi-tools is built on PyTorch and PyTorch Lightning, allowing it to automatically utilize available GPUs for significantly faster model training.

Can I use this for multi-modal data integration?

Yes, the skill includes guidance for models like totalVI and MultiVI, which integrate protein, RNA, and chromatin accessibility data.

Can this skill help with technical batch correction?

Yes, it provides specialized models like scVI and scANVI that are specifically designed to handle technical variation and batch effects in genomic data.

Is it compatible with Scanpy for downstream analysis?

Yes, scvi-tools is designed to be AnnData-centric, making it easy to move between scvi-tools for modeling and Scanpy for visualization or clustering.

scvi-tools for Single-Cell Analysis

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

byricable

•

Data Science & ML

Facilitates advanced probabilistic modeling and analysis of single-cell omics data using deep generative models.

About

This skill empowers Claude to perform complex single-cell genomic analyses using the scvi-tools framework. It provides expert guidance for tasks such as batch correction, dimensionality reduction, and multi-modal integration across various datasets like scRNA-seq and scATAC-seq. By leveraging variational inference and deep learning architectures built on PyTorch, it helps researchers and developers implement robust bioinformatics pipelines, conduct differential expression analysis, and explore spatial transcriptomics with statistically sound methodology and GPU-accelerated efficiency.

Key Features

Comprehensive probabilistic modeling for scRNA-seq, scATAC-seq, and CITE-seq
1 GitHub stars
Seamless integration with AnnData and the Scanpy ecosystem
GPU-accelerated training for large-scale single-cell datasets
Advanced spatial transcriptomics deconvolution and mapping
Automated batch correction and multi-sample data integration

Use Cases

Integrating diverse single-cell datasets to remove technical batch effects
Performing joint analysis of protein and RNA expression in CITE-seq data
Deconvolving cell types in spatial transcriptomics using reference datasets

About

Key Features

Comprehensive probabilistic modeling for scRNA-seq, scATAC-seq, and CITE-seq
1 GitHub stars
Seamless integration with AnnData and the Scanpy ecosystem
GPU-accelerated training for large-scale single-cell datasets
Advanced spatial transcriptomics deconvolution and mapping
Automated batch correction and multi-sample data integration

Use Cases

Integrating diverse single-cell datasets to remove technical batch effects
Performing joint analysis of protein and RNA expression in CITE-seq data
Deconvolving cell types in spatial transcriptomics using reference datasets