PerturbDiff: Functional Diffusion for Single-Cell Perturbation Modeling

TL;DR

PerturbDiff introduces a diffusion-based model that predicts cellular responses to perturbations by modeling entire distributions, effectively capturing variability due to hidden factors and outperforming existing methods.

Contribution

It presents a novel distribution-level diffusion approach for single-cell perturbation modeling, addressing variability from unobservable factors.

Findings

Achieves state-of-the-art response prediction performance.

Generalizes well to unseen perturbations.

Captures population-level response shifts.

Abstract

Building Virtual Cells that can accurately simulate cellular responses to perturbations is a long-standing goal in systems biology. A fundamental challenge is that high-throughput single-cell sequencing is destructive: the same cell cannot be observed both before and after a perturbation. Thus, perturbation prediction requires mapping unpaired control and perturbed populations. Existing models address this by learning maps between distributions, but typically assume a single fixed response distribution when conditioned on observed cellular context (e.g., cell type) and the perturbation type. In reality, responses vary systematically due to unobservable latent factors such as microenvironmental fluctuations and complex batch effects, forming a manifold of possible distributions for the same observed conditions. To account for this variability, we introduce PerturbDiff, which shifts modeling from individual cells to entire distributions. By embedding distributions as points in a Hilbert space, we define a diffusion-based generative process operating directly over probability distributions. This allows PerturbDiff to capture population-level response shifts across hidden factors. Benchmarks on established datasets show that PerturbDiff achieves state-of-the-art performance in single-cell response prediction and generalizes substantially better to unseen perturbations. See our project page (https://katarinayuan.github.io/PerturbDiff-ProjectPage/), where code and data will be made publicly available (https://github.com/DeepGraphLearning/PerturbDiff).