CellForge: Agentic Design of Virtual Cell Models

2025-08-04

AI Summary

Virtual cell modeling faces challenges including high biological system complexity, difficulty integrating multi-source heterogeneous data, and strong reliance on cross-disciplinary expertise. Method: This paper proposes an end-to-end automated modeling framework based on multi-agent collaboration. It employs specialized agents—literature retrieval, task analysis, method design, and code generation—coordinated by a central orchestrator that iteratively negotiates to reach consensus on modeling strategies. The framework synergistically integrates large language models with single-cell multi-omics analytics and deep learning paradigms to enable fully automated translation from raw data to executable predictive code. Contribution/Results: Evaluated on six cross-modal perturbation prediction tasks (e.g., gene knockout, drug treatment), the framework consistently outperforms state-of-the-art methods, achieving significant improvements in both prediction accuracy and generalization capability. All code is publicly released to ensure full reproducibility.

Technology Category

• Multiagent Systems: Modeling other Agents
• Machine Learning: Deep Generative Models & Autoencoders
• Cognitive Modeling & Cognitive Systems: Agent Architectures

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems

Abstract

Virtual cell modeling represents an emerging frontier at the intersection of artificial intelligence and biology, aiming to predict quantities such as responses to diverse perturbations quantitatively. However, autonomously building computational models for virtual cells is challenging due to the complexity of biological systems, the heterogeneity of data modalities, and the need for domain-specific expertise across multiple disciplines. Here, we introduce CellForge, an agentic system that leverages a multi-agent framework that transforms presented biological datasets and research objectives directly into optimized computational models for virtual cells. More specifically, given only raw single-cell multi-omics data and task descriptions as input, CellForge outputs both an optimized model architecture and executable code for training virtual cell models and inference. The framework integrates three core modules: Task Analysis for presented dataset characterization and relevant literature retrieval, Method Design, where specialized agents collaboratively develop optimized modeling strategies, and Experiment Execution for automated generation of code. The agents in the Design module are separated into experts with differing perspectives and a central moderator, and have to collaboratively exchange solutions until they achieve a reasonable consensus. We demonstrate CellForge's capabilities in single-cell perturbation prediction, using six diverse datasets that encompass gene knockouts, drug treatments, and cytokine stimulations across multiple modalities. CellForge consistently outperforms task-specific state-of-the-art methods. Overall, CellForge demonstrates how iterative interaction between LLM agents with differing perspectives provides better solutions than directly addressing a modeling challenge. Our code is publicly available at https://github.com/gersteinlab/CellForge.

Problem

• Autonomously building virtual cell computational models
• Handling biological system complexity and data heterogeneity
• Integrating multi-disciplinary expertise for model optimization

Innovation

• Multi-agent framework for virtual cell modeling
• Automated model architecture and code generation
• Iterative agent collaboration for optimized solutions