Does this skill support GPU acceleration?

Yes, the skill provides instructions for enabling GPU support through PyTorch Lightning to significantly speed up model training on large genomic datasets.

Do I need to normalize my data before using this skill?

No, scvi-tools models are designed to work with raw count data. This skill will guide you through the correct setup using unnormalized counts for optimal performance.

Is it compatible with Scanpy and AnnData?

Yes, the skill focuses on workflows that integrate scvi-tools outputs directly into AnnData objects, making it fully compatible with the Scanpy ecosystem for downstream analysis.

Can I use this for spatial transcriptomics?

Absolutely. The skill includes support for specialized models like DestVI and Stereoscope for spatial deconvolution and mapping tasks.

Single-Cell Genomics Analysis (scvi-tools)

Name: Single-Cell Genomics Analysis (scvi-tools)
Author: plurigrid

byplurigrid

•

Ciencia de Datos y ML

Analyzes single-cell omics data using probabilistic models for tasks like batch correction, dimensionality reduction, and differential expression.

This skill empowers Claude to perform sophisticated single-cell genomic analysis using the scvi-tools framework. It provides expert guidance on implementing deep generative models for scRNA-seq, scATAC-seq, and spatial transcriptomics, ensuring best practices in data registration, model training, and latent space extraction. Whether you are performing cross-sample integration, identifying cell types, or mapping spatial distributions, this skill streamlines the complex workflow of probabilistic modeling in bioinformatics while maintaining seamless compatibility with the Scanpy ecosystem.

Características Principales

018 GitHub stars

02Automated batch correction and multimodal data integration (CITE-seq, Multiome)

03Probabilistic modeling for single-cell transcriptomics and chromatin accessibility

04Spatial transcriptomics deconvolution and cell-environment mapping

05Statistical differential expression analysis using generative models

06Seamless integration with AnnData and Python-based bioinformatics pipelines

Casos de Uso

01Jointly modeling protein and RNA expression in CITE-seq experiments

02Integrating multiple scRNA-seq datasets across different donors or labs

03Performing high-resolution spatial mapping of cell types in tissue samples

Características Principales

018 GitHub stars

02Automated batch correction and multimodal data integration (CITE-seq, Multiome)

03Probabilistic modeling for single-cell transcriptomics and chromatin accessibility

04Spatial transcriptomics deconvolution and cell-environment mapping

05Statistical differential expression analysis using generative models

06Seamless integration with AnnData and Python-based bioinformatics pipelines

Casos de Uso

01Jointly modeling protein and RNA expression in CITE-seq experiments

02Integrating multiple scRNA-seq datasets across different donors or labs

03Performing high-resolution spatial mapping of cell types in tissue samples