arXiv cs.CL· Xiangru Tang, Jiapeng Chen, Daniel Shao, Zhuoyun Yu, Yan Cui, Weixu Wang, Fang Wu, Yuchen Zhuang, Wenqi Shi, Zhi Huang, Arman Cohan, Xihong Lin, Fabian Theis, Smita Krishnaswamy, Mark Gerstein·· 2 天前AI 评分48
CellForge：用 AI 科学家团队构建虚拟细胞模型
AI Scientists for Building Virtual-Cell Models
AI 导读
CellForge 是一支由自主智能体组成的团队，可端到端完成单细胞扰动建模：从文献调研、数据解析到生成模型假设、编写训练代码并跑实验，最终返回含架构、代码与验证结果的预测器，全程无需人工指定设计、损失或训练协议。
正文 · 原文
Authors:Xiangru Tang, Jiapeng Chen, Daniel Shao, Zhuoyun Yu, Yan Cui, Weixu Wang, Fang Wu, Yuchen Zhuang, Wenqi Shi, Zhi Huang, Arman Cohan, Xihong Lin, Fabian Theis, Smita Krishnaswamy, Mark Gerstein
Abstract:Autonomous agents, from virtual laboratories to biological data-analysis agents and co-scientists, have begun to turn scientific goals into hypotheses. Whether they can go further and convert real, noisy and complex biological data into executable model hypotheses and architectures that encode testable assumptions remains unclear. Predicting how cells respond to genetic and chemical perturbations is a natural test of this ability: it is central to decoding disease mechanisms and therapeutic discovery, yet it demands cross-disciplinary expertise rarely concentrated in a single research group. Here, we introduce CellForge, an organized team of autonomous agents for single-cell perturbation modeling. Given a dataset and a task, it surveys the literature, parses the data, identifies relevant baselines, and formulates a model hypothesis through critic-mediated discussion among agents specialized in data modeling, single-cell biology, and deep learning. It then writes the training code, runs the experiments, and returns a trained predictor including architecture, code, and validation results. The architecture itself is the hypothesis: CellForge produces it end-to-end, with no human specifying the design, loss, or training protocol. Across six perturbation datasets spanning transcriptomic, epigenomic, and proteomic contexts, CellForge yields models that are competitive with hand-crafted state-of-the-art results on established tasks and extends the same workflow to paired multi-omic readouts. Blinded human evaluators rate CellForge's research plans as more scientifically sound and innovative than those from single LLMs or deep-research systems. In summary, we show autonomous agents can complete the full hypothesis-to-validation cycle in single-cell biology, with an AI effectively building another AI, in a complex setting that currently requires a team of human experts.
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Quantitative Methods (q-bio.QM) |
| Cite as: | arXiv:2508.02276 [cs.LG] |
| (or arXiv:2508.02276v3 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2508.02276 arXiv-issued DOI via DataCite |
Submission history
From: Xiangru Tang [view email]
[v1]
Mon, 4 Aug 2025 10:43:31 UTC (5,948 KB)
[v2]
Wed, 4 Feb 2026 19:54:36 UTC (8,517 KB)
[v3]
Wed, 14 Oct 2026 04:07:33 UTC (24,994 KB)
来源：arXiv cs.CL · arxiv.org