Train probabilistic generative models on AnnData counts for batch-corrected embeddings.
npx skills add https://github.com/shushuzn/Rairos --skill scvi-tools-shushuzn
Please help me install this Agent Skill. Skill: scvi-tools Source: https://github.com/shushuzn/Rairos/tree/main/skills/scvi-tools Command: npx skills add https://github.com/shushuzn/Rairos --skill scvi-tools-shushuzn
scvi-tools solves the challenge of analyzing noisy, sparse single-cell data by providing probabilistic deep generative models that perform batch correction, integration, and statistically principled downstream inference.
Use the scvi-tools skill to train a generative model on your AnnData counts and extract a batch-corrected latent embedding for downstream UMAP and clustering.
High-intent search queries and answers about installing and using this skill.
Batch correction for single-cell RNA-seq data is performed by training a probabilistic generative model on raw count matrices in AnnData format. This process produces batch-corrected latent embeddings that remain robust across technical variation for downstream clustering and visualization.
Multimodal single-cell integration of RNA and ATAC data is supported through models like MultiVI for joint analysis and totalVI for RNA and protein data. These models require AnnData inputs with raw count-like matrices to generate unified latent representations and normalized outputs.
Probabilistic modeling improves differential expression testing by applying Bayesian inference to compute posterior distributions. This uncertainty-aware approach produces statistically principled differential expression and accessibility results rather than relying on deterministic point estimates.
A GPU is not strictly required to train single-cell models for spatial transcriptomics deconvolution, as training runs on PyTorch backends with optional GPU acceleration. Models like DestVI and Stereoscope can process AnnData inputs on standard compute environments.
The best way to model single-cell ATAC-seq accessibility is using specialized probabilistic models like PeakVI, PoissonVI, or scBasset. These models take raw count-like AnnData matrices to produce denoised accessibility estimates and Bayesian differential accessibility results.
LeonChaoXTrain probabilistic generative models on AnnData counts for single-cell omics analysis.
chenyhvvvvCorrect batch effects across spatial slices using Scanorama integration.
ZurybrGuide MCMC sampling with RStan or PyStan for Bayesian models.
jasrajtulsiAnalyze single-cell and multi-omic datasets with probabilistic deep generative models.
