Analyze single-cell omics data with probabilistic deep generative models.
npx skills add https://github.com/dralkh/seerai --skill scvi-tools-dralkh
Please help me install this Agent Skill. Skill: scvi-tools Source: https://github.com/dralkh/seerai/tree/main/skills/scvi-tools Command: npx skills add https://github.com/dralkh/seerai --skill scvi-tools-dralkh
This Skill helps researchers analyze complex single-cell and spatial omics data with probabilistic deep learning instead of ad hoc preprocessing and manual comparisons.
Ask for the scvi-tools workflow to register raw count data, train the right model for my modality, and return latent embeddings, normalized outputs, and differential analysis results.
High-intent search queries and answers about installing and using this skill.
Batch correction for scRNA-seq data uses probabilistic deep generative models to learn batch-invariant latent spaces. You train on raw count matrices to generate latent embeddings that enable clustering, visualization, and downstream analysis across different batches.
Multimodal integration applies variational inference to align scRNA-seq, ATAC-seq, and cytometry data within a shared latent space. This probabilistic approach replaces ad hoc preprocessing with statistically grounded latent representations for cross-modality analysis.
Spatial transcriptomics deconvolution is supported through probabilistic deep generative models. The workflow processes spatial data to estimate cell type proportions and perform accessibility testing with built-in uncertainty estimates for spatially resolved datasets.
Yes, single-cell differential expression analysis requires raw count-based AnnData or MuData inputs. You must register the count data and set up the specific model before training to generate Bayesian differential expression results.
Semi-supervised cell annotation uses probabilistic models to label cells, map queries to references, and propagate labels across studies. This transfer learning approach leverages trained latent spaces to classify and transfer annotations to new datasets.
Variational inference for trajectory analysis depends on model-specific setup and proper latent-space training. While it handles multiple omics modalities, it requires raw count inputs and careful model selection to avoid inaccurate representations of complex cellular trajectories.
jasrajtulsiAnalyze single-cell and multi-omic datasets with probabilistic deep generative models.
ovachieverAnalyze scRNA-seq data with Scanpy workflows for QC through trajectory inference.
jackspaceModel single-cell omics data with scvi-tools for batch-corrected analysis.
sanand0Analyze single-cell omics data using probabilistic deep generative models.
