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.
Added function custom_predict that allows you to predict the output of the model (latents and the gene expression reconstruction) with respect to only the specific covariate keys, given to the function.
Added function
custom_predictthat allows you to predict the output of the model (latents and the gene expression reconstruction) with respect to only the specific covariate keys, given to the function.