scvi-tools
Probabilistic single-cell models for batch correction and integration.
Install
gh skill install K-Dense-AI/scientific-agent-skills scvi-tools
Batch correction, transfer learning, multimodal or spatial single-cell integration, or differential expression that needs uncertainty.
- Author
- K-Dense Inc.
- License
- BSD-3-Clause
Covers the deep generative model family built on PyTorch: scVI and scANVI for RNA, PeakVI and scBasset for chromatin accessibility, totalVI and MultiVI for multimodal data, DestVI and Stereoscope for spatial deconvolution. A recurring practical trap is handled head on — some models live under scvi.model and others under scvi.external, and the reference files state which namespace each one belongs to. Differential expression here returns uncertainty rather than a bare p-value, which is the main reason to reach for it instead of a standard Scanpy pipeline.
Similar skills
Scanpy
SkillStandard single-cell RNA-seq workflow from QC through clustering and marker genes.
PyDESeq2
SkillRuns bulk RNA-seq differential expression in Python instead of R.
PyMC
SkillBayesian modeling with PyMC: priors, NUTS sampling and model comparison.
Comments(0)
Sign in to comment