arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.00197q-bio.QMcs.AIcs.CL
收藏

Rank-and-Reason: 多智能体协作加速零样本蛋白质突变预测

Rank-and-Reason: Multi-Agent Collaboration Accelerates Zero-Shot Protein Mutation Prediction

Shanghai Jiao Tong University(上海交通大学);Shanghai Innovation Institute(上海创新研究院);Southern University of Science(南方大学(科学));East China University of Science and Technology(东中国科学与技术大学)

Yang Tan, Yuanxi Yu, Can Wu, Bozitao Zhong, Mingchen Li, Guisheng Fan, Jiankang Zhu, Yafeng Liang, Nanqing Dong, Liang Hong

AI总结:

Rank-and-Reason通过多智能体协作提升零样本蛋白质突变预测的准确性与效率,显著提高湿实验验证的成功率。

摘要:

零样本突变预测对于低资源蛋白质工程至关重要,但现有蛋白质语言模型(PLMs)往往产生统计上自信的结果,而忽略了基本的生物物理约束。目前,选择候选者进行湿实验验证依赖于手动专家审查PLM输出,这一过程效率低下、主观性强且高度依赖领域专业知识。为了解决这个问题,我们提出了Rank-and-Reason(VenusRAR),一个两阶段智能体框架,用于自动化此工作流并最大化预期湿实验适应性。在Rank阶段,计算专家和虚拟生物学家聚合了一个上下文感知的多模态集成,建立了一个新的Spearman相关记录为0.551(vs. 0.518)在ProteinGym。在Reason阶段,智能体专家小组采用推理链推理来审查候选者 against 几何和结构约束,使Top-5命中率提高了367%在ProteinGym-DMS99。在Cas12i3核酸酶上的湿实验验证进一步证实了该框架的有效性,实现了46.7%的阳性率,并识别出两个具有4.23倍和5.05倍活性提升的新型突变体。代码和数据集已发布在GitHub(https://github.com/ai4protein/VenusRAR/)。

英文摘要:

Zero-shot mutation prediction is vital for low-resource protein engineering, yet existing protein language models (PLMs) often yield statistically confident results that ignore fundamental biophysical constraints. Currently, selecting candidates for wet-lab validation relies on manual expert auditing of PLM outputs, a process that is inefficient, subjective, and highly dependent on domain expertise. To address this, we propose Rank-and-Reason (VenusRAR), a two-stage agentic framework to automate this workflow and maximize expected wet-lab fitness. In the Rank-Stage, a Computational Expert and Virtual Biologist aggregate a context-aware multi-modal ensemble, establishing a new Spearman correlation record of 0.551 (vs. 0.518) on ProteinGym. In the Reason-Stage, an agentic Expert Panel employs chain-of-thought reasoning to audit candidates against geometric and structural constraints, improving the Top-5 Hit Rate by up to 367% on ProteinGym-DMS99. The wet-lab validation on Cas12i3 nuclease further confirms the framework's efficacy, achieving a 46.7% positive rate and identifying two novel mutants with 4.23-fold and 5.05-fold activity improvements. Code and datasets are released on GitHub (https://github.com/ai4protein/VenusRAR/).

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑