Atlas
namespace/model
Sign in
Models / single-cell / cpa

single-cell/cpa

human-cell-atlasperturbationscrna-seqsingle-cellsingle-cell-genomicssingle-cell-rna-seqsource-code
Visibility: PUBLIC Updated 8/7/2026
Model cardFiles Benchmarks

Overview

The Compositional Perturbation Autoencoder (CPA) is a deep generative framework to learn effects of perturbations at the single-cell level. CPA performs OOD predictions of unseen combinations of drugs, learns interpretable embeddings, estimates dose-response curves, and provides uncertainty estimates.

Files

Upload weights or artifacts — large files transfer resumably (TUS) and land as Git LFS objects in this model's repo.