VCLMU: Mechanism-Centric Virtual Cell World Modeling for Perturbation Response

Yuwei Miao; Azim Dehghani Amirabad; Scott Oloff; Junzhou Huang; Tianyu Cui; Rui Liao

cs.AIOct 3, 2026

VCLMU: Mechanism-Centric Virtual Cell World Modeling for Perturbation Response

Authors: Yuwei Miao, Azim Dehghani Amirabad, Scott Oloff, Junzhou Huang, Tianyu Cui, Rui Liao

Organizations: Johnson & Johnson Innovative Medicine · University of Texas at Arlington

Abstract

Predicting cellular responses to genetic perturbations is a central capability for virtual cells and a key step toward computational modeling of biological interventions. Most existing models directly map an unperturbed molecular profile and perturba- tion to the resulting observation without explicitly representing the latent cellular transition induced by the intervention. We introduce a mechanism-centric virtual cell world model that represents cellular state as a set of Latent Mechanism Units (LMUs) and treats genetic perturbations as actions on these latent states. Each LMU combines a reusable identity grounded in multimodal biological evidence with an observation-specific state, allowing a perturbation to induce mechanism- specific stochastic transitions before decoding the resulting transcriptional response. We train VCLMU through two-stage pretraining, first on around 200K pseudo-bulk perturbation profiles and then on gene-aligned single-cell perturbation data. Across six perturbation-disjoint benchmarks, VCLMU consistently improves perturbation- specific response recovery over strong baselines while maintaining competitive global response accuracy. We further analyze learned LMUs through enrichment between perturbation responses and LMU gene sets and show that they capture structured biological response programs. These results support mechanism-level latent state transition as a useful formulation for virtual cell models that aim to predict and interpret cellular responses to biological interventions.

Figures & tables

Figure 1: Overview of the mechanism-centric virtual cell world model. Multimodal biological evidence is composed into reusable Latent Mechanism Unit (LMU) identities. A control profile

\mathbf{x}_{i}^{c}

is encoded and routed to a sparse active LMU set with control mechanism states

\mathbf{H}_{i}^{c}

, while the perturbation action

a_{i}

is encoded as

\boldsymbol{\psi}_{i}

. Each active LMU undergoes an action-conditioned stochastic transition to produce predicted perturbed mechanism states, which are decoded into the response distribution

p(\Delta\mathbf{x}_{i}\mid\mathbf{x}_{i}^{c},a_{i})

and the predicted profile

\widehat{\mathbf{x}}_{i}^{p}=\mathbf{x}_{i}^{c}+\widehat{\boldsymbol{\mu}}_{i}

. During training, the observed perturbed profile is encoded through the same active LMUs to provide transition targets and support inverse prediction of the perturbation, without entering the forward prediction path.

Figure 2: Perturbation prediction and LMU interpretability. (a) Jaccard@20 across six evaluation datasets for VCLMU-FT and baselines. (b) Stage 1 versus Stage 2 pretraining under the same downstream fine-tuning protocol; annotations indicate the Jaccard@20 gain. (c) $\Delta$ PCC versus $\Delta$ PCC@20, showing global and DE-focused Pearson correlation jointly. (d) Enrichment of perturbation-specific response genes in learned LMU candidate gene sets across held-out perturbations after excluding ubiquitous response genes and subtracting the shared response; color indicates $-\log_{10}(\mathrm{BH\text{-}FDR})$ . (e) Six representative significant associations between perturbations and LMUs whose overlapping response genes form biologically coherent programs spanning interferon response, mitotic regulation, T-cell signaling, T-cell activation, oxidative phosphorylation, and RNA processing; numbers indicate overlapping response genes.

Appendix figures & tables2 assets

Supplementary material from the paper’s appendix.

Appendix

Symbol	Meaning
$G$ , $g$	number of genes; gene index
$K$ , $k$	number of LMUs in the bank; LMU index
$K_{a}$ , $\mathcal{A}_{i}$	number of active LMUs for sample $i$ ; their index set
$C$ , $\mathcal{C}_{k}$	candidate genes per LMU; candidate set of LMU $k$
$d$ , $d_{z}$	model width; latent transition width
$\mathbf{x}_{i}^{c}$ , $\mathbf{x}_{i}^{p}$ , $\widehat{\mathbf{x}}_{i}^{p}$	control profile; observed perturbed profile; predicted perturbed profile

Appendix

Table 1: Notation used in the main text and appendix.

Evidence	Source	Description	Dim.
$m_{1}$	Training co-expression	Principal components of the training-set gene–gene co-expression matrix, capturing major transcriptional covariance patterns.	$\leq 256$
$m_{2}$	Gene Ontology [ 35 ]	Gene–term association matrix compressed using truncated SVD to summarize biological process, molecular function, and cellular component annotations.	512
$m_{3}$	MSigDB Hallmark [ 36 ]	Binary membership in the Hallmark gene sets, representing approximately 50 curated biological programs.	$\sim 50$
$m_{4}$	PROGENy [ 28 ]	Association with the 14 PROGENy signaling pathways, providing pathway-level functional information.	14
$m_{5}$	Gene2Vec [ 37 ]	Pretrained gene embeddings capturing functional similarity learned from large-scale biological co-occurrence.	200
$m_{6}$	STRING + Mashup [ 38 , 39 ]	Protein interaction network embedding capturing both local and global network topology.	800

Appendix

Table 2: Biological evidence representations used to construct the multimodal gene evidence embedding.

Single-Cell Perturbation Modeling

Single-cell perturbation modeling predicts how genetic or chemical interventions change cellular states across cells and contexts.

See all 61 papers

Lingshu-Cell: A generative cellular world model for transcriptome modeling toward virtual cells
Mar 26, 2026Han Zhang, Guo-Hua Yuan, Chaohao Yuan +6Single-Cell Perturbation Modeling Single-Cell Transcriptomics
DAMO Academy, Alibaba Group
AROMA: Augmented Reasoning Over a Multimodal Architecture for Virtual Cell Genetic Perturbation Modeling
Apr 22, 2026Zhenyu Wang, Geyan Ye, Wei Liu +1Multimodal Graph Learning
AI for Life Sciences Lab, Tencent, Shenzhen, China
PerturbDiff: Functional Diffusion for Single-Cell Perturbation Modeling
Feb 23, 2026Xinyu Yuan, Xixian Liu, Ya Shi Zhang +3Single-Cell Perturbation Modeling Single-Cell Transcriptomics
Mila - Québec AI Institute · University of Montréal · University of Ottawa +3

Mar 26, 2026J/K move · Enter open · S save

Lingshu-Cell: A generative cellular world model for transcriptome modeling toward virtual cells

Han Zhang, Guo-Hua Yuan, Chaohao Yuan, Tingyang Xu+5

DAMO Academy, Alibaba Group

Single-Cell Perturbation Modeling Single-Cell Transcriptomics World Models Diffusion Models

Open paper AskPDF arXiv

Modeling cellular states and predicting their responses to perturbations are central challenges in computational biology and the development of virtual cells. Existing foundation models for single-cell transcriptomics provide powerful static representations, but they do not explicitly model the distribution of cellular states for generative simulation. Here, we introduce Lingshu-Cell, a masked discrete diffusion model that learns transcriptomic state distributions and supports conditional simulation under perturbation. By operating directly in a discrete token space that is compatible with the sparse, non-sequential nature of single-cell transcriptomic data, Lingshu-Cell captures complex transcriptome-wide expression dependencies across approximately 18,000 genes without relying on prior gene selection, such as filtering by high variability or ranking by expression level. Across diverse tissues and species, Lingshu-Cell accurately reproduces transcriptomic distributions, marker-gene expression patterns and cell-subtype proportions, demonstrating its ability to capture complex cellular heterogeneity. Moreover, by jointly embedding cell type or donor identity with perturbation, Lingshu-Cell can predict whole-transcriptome expression changes for novel combinations of identity and perturbation. It achieves leading performance on the Virtual Cell Challenge H1 genetic perturbation benchmark and in predicting cytokine-induced responses in human PBMCs. Together, these results establish Lingshu-Cell as a flexible cellular world model for in silico simulation of cell states and perturbation responses, laying the foundation for a new paradigm in biological discovery and perturbation screening.

Abstract

Single-Cell Perturbation Modeling

Explore similar work

Lingshu-Cell: A generative cellular world model for transcriptome modeling toward virtual cells

AROMA: Augmented Reasoning Over a Multimodal Architecture for Virtual Cell Genetic Perturbation Modeling

PerturbDiff: Functional Diffusion for Single-Cell Perturbation Modeling