scvi-tools documentation

scvi-tools (single-cell variational inference tools) is a package for end-to-end analysis of single-cell omics data. The package is primarily developed and maintained by the Yosef Lab at UC Berkeley and is composed of several deep generative models for omics data analysis:

  • scVI for analysis of single-cell RNA-seq data [Lopez18]

  • scANVI for cell annotation of scRNA-seq data using semi-labeled examples [Xu19]

  • totalVI for analysis of CITE-seq data [GayosoSteier20]

  • gimVI for imputation of missing genes in spatial transcriptomics from scRNA-seq data [Lopez19]

  • AutoZI for assessing gene-specific levels of zero-inflation in scRNA-seq data [Clivio19]

  • LDVAE for an interpretable linear factor model version of scVI [Svensson20]

These models are able to simultaneously perform many downstream tasks such as learning low-dimensional cell representations, harmonizing datasets from different experiments, and identifying differential expressed features [Boyeau19]. By levaraging advances in stochastic optimization, these models scale to millions of cells.

  • If you find a model useful for your research, please consider citing the corresponding publication.

Important

scvi is now scvi-tools. If you’d like to view documentation for scvi, please change the documentation version using the menu at the bottom right (versions <= 0.6.8).

getting started with scvi action icon
Getting started

New to scvi-tools? Check out the installation guide.

scvi user guide action icon
User guide

The tutorials provide in-depth information on running scvi-tools models.

api of scvi action icon
API reference

The API reference contains a detailed description of the scvi-tools API.

contribute to scvi-tools action icon
Developer guide

Want to improve scvi-tools? The contributing guidelines will guide you through the process.