Deep Learning

CellDISECT
CellDISECT (Cell DISentangled Experts for Covariate counTerfactuals) is a powerful causal generative model that enhances single-cell analysis by disentangling variations, making counterfactual predictions, and achieving flexible fairness.
Code Code
CellDISECT
Integrating multi-covariate disentanglement with counterfactual analysis on synthetic data enables cell type discovery and counterfactual predictions
CellDISECT models how cell states change across covariates and perturbations, making it easier to ask counterfactual questions in single-cell biology.
Stathis Megas, Arian Amani, Antony Rose, Olli Dufva, Kian Shamsaie, Hesam Asadollahzadeh, Krzysztof Polanski, Muzlifah Haniffa, Sarah Amalia Teichmann, Mohammad Lotfollahi
PDF Cite Code Project DOI bioRxiv
Integrating multi-covariate disentanglement with counterfactual analysis on synthetic data enables cell type discovery and counterfactual predictions
Leveraging Machine Learning to Predict Cellular Behavior in Drug Treatments
Why is ML important in Drug Discovery?
Arian Amani
Sep 2, 2024 1 min read Blog Posts
Leveraging Machine Learning to Predict Cellular Behavior in Drug Treatments
A Deep Learning Road Map And Where To Start
A Deep Learning Road Map And Where To Start
Arian Amani
Sep 2, 2022 10 min read Blog Posts
A Deep Learning Road Map And Where To Start
CPA (Compositional Perturbation Autoencoder)
CPA is a deep generative framework to learn effects of perturbations at the single-cell level. It performs OOD predictions of unseen combinations of drugs, learns interpretable embeddings, estimates dose-response curves, and provides uncertainty estimates.
Code Code
CPA (Compositional Perturbation Autoencoder)