DRIFT: Disentangled Responsive-Invariant Flow Transport for Single-Cell Perturbation Prediction

2026-09-28

AI Summary

This study addresses the challenge of confounding between intrinsic heterogeneity and perturbation effects in single-cell perturbation prediction by proposing a unified framework that integrates disentangled representations with conditional flow matching. Specifically, the method employs a variational autoencoder augmented with information-theoretic invariance constraints to decouple cellular states into invariant and response features. Conditional flow matching transport is then applied exclusively to the response component, thereby overcoming the limitations of predefined mechanisms and enabling confounding-free modeling of perturbation effects. Experimental results demonstrate that the proposed approach surpasses existing state-of-the-art methods on benchmarks for both combinatorial and unseen perturbation prediction.

Abstract

Predicting cellular responses to perturbations is a central problem in cellular biology, with broad applications in systems biology and drug discovery. This task is challenging because cellular responses can be complex and cell-state dependent, intrinsic cell-to-cell variability can be confounded with perturbation effects, and destructive single-cell RNA sequencing precludes paired measurements of the same cell before and after treatment. Flow matching transports control cells to perturbed states flexibly, but acting on the full cell state can confound perturbation effects with pre-existing cell-to-cell variability. Disentangled approaches separate responsive from invariant components, but model perturbations through prescribed mechanisms, such as latent shifts or graph edits, limiting their flexibility. We address both limitations in a unified framework. A variational encoder disentangles each cell into an invariant block, capturing state unaffected by the perturbation, and a responsive block, capturing state it changes, through conditional priors and an information-theoretic invariance constraint. Conditional flow matching transports only the responsive block, conditioned on the perturbation and invariant state, yielding a flexible, data-driven model of perturbation effects without confounding pre-existing variability. Across several benchmarks, our method outperforms the strongest published method in settings involving combinatorial and unseen perturbation prediction.

Problem

• single-cell perturbation prediction
• disentangled representation
• cell-to-cell variability
• flow matching

Innovation

• Disentangled Representation
• Flow Matching
• Single-Cell Perturbation Prediction
• Variational Encoder
• Information-Theoretic Constraint