Modelling Cellular Perturbations with the Sparse Additive Mechanism Shift Variational Autoencoder

2023/11/05 by Michael Bereket, Bereket, Michael, Theofanis Karaletsos +1 · 10 citations
Biochemistry, Genetics and Molecular Biology · Computer Science · #Artificial Intelligence (cs.AI) #Cell Image Analysis Techniques #Explainable Artificial Intelligence (XAI) #FOS: Biological sciences #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Quantitative Methods (q-bio.QM) #Single-cell and spatial transcriptomics

paper · pdf · doi:10.48550/arxiv.2311.02794

openalex publication_date 2023/11/05 · openalex created_date 2023/11/08 · openalex updated_date 2026/07/28

Abstract

Generative models of observations under interventions have been a vibrant topic of interest across machine learning and the sciences in recent years. For example, in drug discovery, there is a need to model the effects of diverse interventions on cells in order to characterize unknown biological mechanisms of action. We propose the Sparse Additive Mechanism Shift Variational Autoencoder, SAMS-VAE, to combine compositionality, disentanglement, and interpretability for perturbation models. SAMS-VAE models the latent state of a perturbed sample as the sum of a local latent variable capturing sample-specific variation and sparse global variables of latent intervention effects. Crucially, SAMS-VAE sparsifies these global latent variables for individual perturbations to identify disentangled, perturbation-specific latent subspaces that are flexibly composable. We evaluate SAMS-VAE both quantitatively and qualitatively on a range of tasks using two popular single cell sequencing datasets. In order to measure perturbation-specific model-properties, we also introduce a framework for evaluation of perturbation models based on average treatment effects with links to posterior predictive checks. SAMS-VAE outperforms comparable models in terms of generalization across in-distribution and out-of-distribution tasks, including a combinatorial reasoning task under resource paucity, and yields interpretable latent structures which correlate strongly to known biological mechanisms. Our results suggest SAMS-VAE is an interesting addition to the modeling toolkit for machine learning-driven scientific discovery.

Cited by

• PerturbPFN: Probing the Limits of Synthetic Priors in Drug Perturbation Modelling
• Departures: Distributional Transport for Single-Cell Perturbation Prediction with Neural Schrödinger Bridges
• Scalable Single-Cell Gene Expression Generation with Latent Diffusion Models
• PRESCRIBE: Predicting Single-Cell Responses with Bayesian Estimation
• MorphGen: Controllable and Morphologically Plausible Generative Cell-Imaging
• SynthPert: Enhancing LLM Biological Reasoning via Synthetic Reasoning Traces for Cellular Perturbation Prediction
• Toward a foundation model of causal cell and tissue biology with a Perturbation Cell and Tissue Atlas
• Modeling Gene Expression Distributional Shifts for Unseen Genetic Perturbations
• GPerturb: Gaussian process modelling of single-cell perturbation data
• Unlasting: Unpaired Single-Cell Multi-Perturbation Estimation by Dual Conditional Diffusion Implicit Bridges

Related

• Learning Causal Representations of Single Cells via Sparse Mechanism Shift Modeling
• CRADLE-VAE: Enhancing Single-Cell Gene Perturbation Modeling with Counterfactual Reasoning-based Artifact Disentanglement
• Generative Intervention Models for Causal Perturbation Modeling
• Exploring the Latent Space of Autoencoders with Interventional Assays
• Supervised Vector Quantized Variational Autoencoder for Learning Interpretable Global Representations