StateXDiff: Cell State-Contextualized Multimodal Diffusion for Single-Cell Perturbation Prediction

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Shi, Peiting, Que, Ningfeng, Huang, Xianzhe, Wang, Xiaofei, Xi, Jianzhong Jeff
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918504320466944
author Shi, Peiting
Que, Ningfeng
Huang, Xianzhe
Wang, Xiaofei
Xi, Jianzhong Jeff
author_facet Shi, Peiting
Que, Ningfeng
Huang, Xianzhe
Wang, Xiaofei
Xi, Jianzhong Jeff
contents Predicting drug-induced cellular state changes at single-cell resolution remains a central challenge in virtual cell modeling, particularly under out-of-distribution (OOD) conditions. Current approaches predominantly rely on RNA-based assays, which often fail to adequately capture the diverse cellular states underlying drug responses. Moreover, conditional distribution shifts and low signal-to-noise ratios frequently cause models to learn spurious correlations rather than genuine state transitions. To address these limitations, we introduce StateXDiff, a cell State-contextualized multimodal (X) Diffusion framework for predicting single-cell responses to drug perturbations. The framework operates sequentially: first, it learns a disentangled, multimodal representation of cellular state by integrating transcriptomic profiles with inferred protein features; second, it employs a conditional diffusion model to generate perturbation-specific changes. Our approach introduces a Virtual Multimodal Cell State, which augments RNA-based representations with protein-level context, and a Mechanism-aware Drug-Gene Template, which consolidates multi-source biological knowledge for accurate drug representation. Generation is driven by a latent-space diffusion Transformer, regularized through quality-aware triplet constraints, including positive drug-protein pairs or protein-drug mismatched pairs, and explicit protein-reliability weighting. Extensive evaluation demonstrates that StateXDiff consistently enhances generalization performance across three challenging settings: unseen cell lines, unseen drugs, and combinatorial perturbations.
format Preprint
id arxiv_https___arxiv_org_abs_2605_16104
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle StateXDiff: Cell State-Contextualized Multimodal Diffusion for Single-Cell Perturbation Prediction
Shi, Peiting
Que, Ningfeng
Huang, Xianzhe
Wang, Xiaofei
Xi, Jianzhong Jeff
Genomics
Quantitative Methods
Predicting drug-induced cellular state changes at single-cell resolution remains a central challenge in virtual cell modeling, particularly under out-of-distribution (OOD) conditions. Current approaches predominantly rely on RNA-based assays, which often fail to adequately capture the diverse cellular states underlying drug responses. Moreover, conditional distribution shifts and low signal-to-noise ratios frequently cause models to learn spurious correlations rather than genuine state transitions. To address these limitations, we introduce StateXDiff, a cell State-contextualized multimodal (X) Diffusion framework for predicting single-cell responses to drug perturbations. The framework operates sequentially: first, it learns a disentangled, multimodal representation of cellular state by integrating transcriptomic profiles with inferred protein features; second, it employs a conditional diffusion model to generate perturbation-specific changes. Our approach introduces a Virtual Multimodal Cell State, which augments RNA-based representations with protein-level context, and a Mechanism-aware Drug-Gene Template, which consolidates multi-source biological knowledge for accurate drug representation. Generation is driven by a latent-space diffusion Transformer, regularized through quality-aware triplet constraints, including positive drug-protein pairs or protein-drug mismatched pairs, and explicit protein-reliability weighting. Extensive evaluation demonstrates that StateXDiff consistently enhances generalization performance across three challenging settings: unseen cell lines, unseen drugs, and combinatorial perturbations.
title StateXDiff: Cell State-Contextualized Multimodal Diffusion for Single-Cell Perturbation Prediction
topic Genomics
Quantitative Methods
url https://arxiv.org/abs/2605.16104