Which omics modalities are supported by this skill?

It supports a wide range of modalities including scRNA-seq, scATAC-seq, CITE-seq (proteins), spatial transcriptomics, methylation, and flow/mass cytometry.

Does it support spatial transcriptomics?

Absolutely. It includes patterns for spatial models like DestVI and Stereoscope for deconvolution and spatial mapping tasks.

Can this skill help with batch effect removal?

Yes, it provides specific patterns for registering batch covariates during the AnnData setup, allowing the models to learn latent representations that are corrected for technical variations.

Do I need to normalize my data before using these models?

No, one of the key requirements for scvi-tools is the use of raw, unnormalized count data, as the probabilistic models account for library size and technical noise internally.

What is the scvi-tools skill for Claude Code?

The scvi-tools skill is a specialized capability that provides guidance and implementation patterns for using the scvi-tools Python library to analyze single-cell genomics data via deep learning.

scvi-tools for Single-Cell Analysis

Name: scvi-tools for Single-Cell Analysis
Author: x-cmd

byx-cmd

•

데이터 과학 및 ML

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

scvi-tools provides a comprehensive Python framework for applying probabilistic models to single-cell genomics. This skill streamlines the implementation of deep learning architectures like scVI, scANVI, and TotalVI, enabling researchers to perform statistically rigorous dimensionality reduction, batch correction, and differential expression analysis. Built on PyTorch and AnnData, it offers a unified API for processing diverse modalities including scRNA-seq, scATAC-seq, and spatial transcriptomics, ensuring scalable and reproducible bioinformatics workflows within the Claude Code environment.

주요 기능

01Probabilistic batch correction and multi-study data integration

02Spatial transcriptomics deconvolution and cell-type mapping

03GPU-accelerated training and inference for large-scale datasets

048 GitHub stars

05Deep generative modeling for scRNA-seq, ATAC-seq, and CITE-seq

06Automated AnnData setup and technical covariate registration

사용 사례

01Identifying cell types through semi-supervised annotation and transfer learning

02Integrating heterogeneous single-cell datasets from multiple studies and technical batches

03Performing statistically rigorous differential expression directly on raw count data

주요 기능

01Probabilistic batch correction and multi-study data integration

02Spatial transcriptomics deconvolution and cell-type mapping

03GPU-accelerated training and inference for large-scale datasets

048 GitHub stars

05Deep generative modeling for scRNA-seq, ATAC-seq, and CITE-seq

06Automated AnnData setup and technical covariate registration

사용 사례

01Identifying cell types through semi-supervised annotation and transfer learning

02Integrating heterogeneous single-cell datasets from multiple studies and technical batches

03Performing statistically rigorous differential expression directly on raw count data