skills.rest
tony-zhelonkin
tony-zhelonkinCOMMUNITY
Explore all skills by this author →
𝕏@TonyZhelonkinGitHub

scvi-contrastivevi

Separate perturbation-specific from background variation in single-cell datasets.

1|Updated Nov 20, 2025
Search Tags:#single-cell#scvi-tools#differential-expression#latent-space#perturb-seq#contrastivevi
One-click install
npx skills add https://github.com/tony-zhelonkin/SciAgent-toolkit --skill scvi-contrastivevi
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: scvi-contrastivevi
Source: https://github.com/tony-zhelonkin/SciAgent-toolkit/tree/main/skills/scvi-contrastivevi
Command: npx skills add https://github.com/tony-zhelonkin/SciAgent-toolkit --skill scvi-contrastivevi

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

ContrastiveVI enables analysis of perturbation experiments by separating salient (perturbation-specific) variation from background (shared) variation in single-cell data.

Core Features & Use Cases

  • Identify target vs background effects in Perturb-seq and related experiments.
  • Train models with dual latent spaces to visualize perturbation effects and adjust for confounders.
  • Use differential expression analysis focused on perturbation-specific signals.

Quick Start

Identify target and background cells, set up an AnnData object with counts, initialize ContrastiveVI, train with n_latent and n_salient, and extract latent representations for downstream analysis.

Frequently Asked Questions about scvi-contrastivevi

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I isolate perturbation-specific signals in single-cell Perturb-seq data?▼

You can isolate perturbation-specific signals in single-cell data by using ContrastiveVI, which separates salient perturbation-specific variation from shared background variation in raw counts datasets.

What is the best way to analyze drug-treatment effects versus background variation in single-cell datasets?▼

Analyzing drug-treatment effects against background variation is done by training ContrastiveVI with dual latent spaces, allowing you to visualize perturbation effects while adjusting for shared confounders.

How do I set up AnnData objects for contrastive analysis of disease versus healthy cells?▼

To set up AnnData objects for contrastive analysis, load your raw counts and call ContrastiveVI.setup_anndata, ensuring target and background cells are properly identified for model initialization.

Can I perform differential expression focused on perturbation-specific signals in single-cell data?▼

Yes, you can perform differential expression on the salient latent space extracted after training ContrastiveVI, focusing strictly on perturbation-specific signals rather than shared background variation.

Do I need to configure specific latent space dimensions for single-cell perturbation analysis?▼

Yes, training ContrastiveVI requires configuring both n_latent for the total latent space and n_salient for the perturbation-specific dimensions to accurately separate signals in single-cell data.

When should I separate salient and background variation instead of using standard single-cell analysis?▼

You should separate salient and background variation when working with Perturb-seq, drug-treatment studies, or disease-versus-healthy comparisons requiring target-versus-background analysis on raw counts.

Related Skills

Explore More Data & Analytics Skills→
StarlitnightlyStarlitnightly
13

omicverse-single-cell-differential-expression

Compare conditions per cell type in AnnData with Wilcoxon, t-test, or memento backends.

Community
Advanced
ya-wayya-way

bio-single-cell-perturb-seq

Identify how pooled genetic perturbations influence transcriptional programs in single cells.

Community
Advanced
tony-zhelonkintony-zhelonkin
1

scvi-mrvi

Model sample identity as biological variation in multi-donor scRNA-seq data with MRVI.

Community
Advanced
nahisahonahisaho
3

scientific-perturbation-analysis

Identify perturbation-driven changes in single-cell gene expression across treatments.

Community
Advanced