Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

VCWorld: A Biological World Model for Virtual Cell Simulation

Zhijian Wei, Runze Ma, Zichen Wang, Zhongmin Li, Shuotong Song, Shuangjia Zheng

TL;DR

VCWorld presents a data-efficient, white-box approach to predicting cellular responses by integrating structured biological knowledge with LLM-based reasoning. It reframes perturbation prediction as gene-centric tasks and employs an open-world knowledge graph with retrieval-augmented, chain-of-thought reasoning to produce interpretable predictions and mechanistic hypotheses. The GeneTAK benchmark enables robust evaluation of gene-level perturbation effects, with VCWorld achieving state-of-the-art performance on DE and DIR tasks and showing strong case-study alignment with biological evidence. The work emphasizes interpretability and biological grounding, offering a path toward credible in silico cellular perturbation modeling and hypothesis generation.

Abstract

Virtual cell modeling aims to predict cellular responses to perturbations. Existing virtual cell models rely heavily on large-scale single-cell datasets, learning explicit mappings between gene expression and perturbations. Although recent models attempt to incorporate multi-source biological information, their generalization remains constrained by data quality, coverage, and batch effects. More critically, these models often function as black boxes, offering predictions without interpretability or consistency with biological principles, which undermines their credibility in scientific research. To address these challenges, we present VCWorld, a cell-level white-box simulator that integrates structured biological knowledge with the iterative reasoning capabilities of large language models to instantiate a biological world model. VCWorld operates in a data-efficient manner to reproduce perturbation-induced signaling cascades and generates interpretable, stepwise predictions alongside explicit mechanistic hypotheses. In drug perturbation benchmarks, VCWorld achieves state-of-the-art predictive performance, and the inferred mechanistic pathways are consistent with publicly available biological evidence.

VCWorld: A Biological World Model for Virtual Cell Simulation

TL;DR

VCWorld presents a data-efficient, white-box approach to predicting cellular responses by integrating structured biological knowledge with LLM-based reasoning. It reframes perturbation prediction as gene-centric tasks and employs an open-world knowledge graph with retrieval-augmented, chain-of-thought reasoning to produce interpretable predictions and mechanistic hypotheses. The GeneTAK benchmark enables robust evaluation of gene-level perturbation effects, with VCWorld achieving state-of-the-art performance on DE and DIR tasks and showing strong case-study alignment with biological evidence. The work emphasizes interpretability and biological grounding, offering a path toward credible in silico cellular perturbation modeling and hypothesis generation.

Abstract

Virtual cell modeling aims to predict cellular responses to perturbations. Existing virtual cell models rely heavily on large-scale single-cell datasets, learning explicit mappings between gene expression and perturbations. Although recent models attempt to incorporate multi-source biological information, their generalization remains constrained by data quality, coverage, and batch effects. More critically, these models often function as black boxes, offering predictions without interpretability or consistency with biological principles, which undermines their credibility in scientific research. To address these challenges, we present VCWorld, a cell-level white-box simulator that integrates structured biological knowledge with the iterative reasoning capabilities of large language models to instantiate a biological world model. VCWorld operates in a data-efficient manner to reproduce perturbation-induced signaling cascades and generates interpretable, stepwise predictions alongside explicit mechanistic hypotheses. In drug perturbation benchmarks, VCWorld achieves state-of-the-art predictive performance, and the inferred mechanistic pathways are consistent with publicly available biological evidence.

Paper Structure

This paper contains 51 sections, 20 equations, 11 figures, 10 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Virtual Cell
  4. LLMs for Biological Reasoning and Prediction
  5. Pert-seq Data Representation
  6. Method
  7. Task Formulation
  8. GeneTAK
  9. Dataset Curation.
  10. VCWorld
  11. Construction of the Open-World Biological Knowledge Graph
  12. Generative Node Feature Representation via LLMs
  13. Graph-guided causal evidence framework
  14. Evidence Synthesis Chain-of-Thought Reasoning
  15. Experiment
  16. ...and 36 more sections

Figures (11)

  • Figure 1: The pipeline of VCWorld. For a given query, VCWorld first retrieves multi-modal biological context from its integrated knowledge base, including pathway information, drug properties, and similar experimental samples. This context is then used to prompt a LLM, which is used to analyze mechanisms and infer how a specific gene will respond. The final output is a prediction for tasks such as differential expression and directional change.
  • Figure 2: (a) Number of single-cell profiles per cell line. (b) Label distributions for the Differential Expression (DE) and Directional Change (DIR) tasks. (c) Schematic for identifying highly variable genes by comparing perturbed cells against a control group.
  • Figure 3: The ablation experiments of VCWorld.
  • Figure 4: VCWorld Inference on (Larotrectinib,MKI67). The schematic illustrates the inference process for Larotrectinib-induced MKI67 downregulation, integrating direct target inference with retrieved biological evidence (BioContext). The LLM synthesizes these inputs via Chain-of-Thought (CoT) reasoning to predict causal relationships.
  • Figure :
  • ...and 6 more figures