Analyzes single-cell omics data using deep generative models and probabilistic frameworks for genomics research.
scvi-tools is a specialized framework designed for the probabilistic modeling of single-cell genomics data, providing robust tools for tasks like dimensionality reduction, batch correction, and multimodal integration. Built on PyTorch, it enables researchers and developers to apply sophisticated deep learning models to scRNA-seq, scATAC-seq, and spatial transcriptomics datasets with a unified API. This skill streamlines the entire analysis pipeline from data preprocessing and covariate registration to differential expression and cell type annotation, ensuring best practices like raw count utilization and GPU acceleration are maintained throughout the workflow.
주요 기능
015 GitHub stars
02Probabilistic differential expression analysis
03Multimodal analysis for CITE-seq and spatial transcriptomics
04Automated batch correction and multi-study data integration
05Seamless integration with the scanpy and AnnData ecosystem
06Deep generative modeling for single-cell genomics
사용 사례
01Conducting semi-supervised cell type annotation for large-scale tissue atlases
02Performing spatial transcriptomics deconvolution and cell-type mapping
03Integrating multiple single-cell datasets across different technical batches or studies
