scDFM: Distributional Flow Matching Model for Robust Single-Cell Perturbation Prediction

TL;DR

scDFM introduces a distributional flow matching model with a novel architecture to predict cellular responses to perturbations, effectively handling noise and population shifts in single-cell data.

Contribution

The paper presents a new generative framework, scDFM, that models full cell population distributions and incorporates a perturbation-aware transformer for improved robustness and accuracy.

Findings

Outperforms prior methods on multiple benchmarks.

Reduces mean squared error by 19.6% in combinatorial settings.

Demonstrates strong generalization to unseen perturbations.

Abstract

A central goal in systems biology and drug discovery is to predict the transcriptional response of cells to perturbations. This task is challenging due to the noisy and sparse nature of single-cell measurements, as well as the fact that perturbations often induce population-level shifts rather than changes in individual cells. Existing deep learning methods typically assume cell-level correspondences, limiting their ability to capture such global effects. We present scDFM, a generative framework based on conditional flow matching that models the full distribution of perturbed cells conditioned on control states. By incorporating a maximum mean discrepancy (MMD) objective, our method aligns perturbed and control populations beyond cell-level correspondences. To further improve robustness to sparsity and noise, we introduce the Perturbation-Aware Differential Transformer (PAD-Transformer), a backbone architecture that leverages gene interaction graphs and differential attention to capture context-specific expression changes. Across multiple genetic and drug perturbation benchmarks, scDFM consistently outperforms prior methods, demonstrating strong generalization in both unseen and combinatorial settings. In the combinatorial setting, it reduces mean squared error by 19.6% relative to the strongest baseline. These results highlight the importance of distribution-level generative modeling for robust in silico perturbation prediction. The code is available at https://github.com/AI4Science-WestlakeU/scDFM