Method predicts cell changes under treatments separating noise from effects

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

Machine Learning

Summary

Predicting how individual cells react to treatments is hard because cells naturally vary and experiments can't track the same cell before and after treatment. The authors created a method that splits each cell's state into parts that change with treatment and parts that stay the same. Their approach moves only the changing part to predict how a treatment will affect cells, avoiding confusion with natural differences. This leads to better predictions, even for combinations of treatments or new treatments not seen before.

What this means in practice

Authors

Mustapha Bounoua, Giulio Franzese, Pietro Michiardi

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.