用多智能体协作自动预测蛋白质突变，提升实验成功率。
Rank-and-Reason: Multi-Agent Collaboration Accelerates Zero-Shot Protein Mutation Prediction
- 两阶段框架：先排序候选突变，再用推理审计结构合理性。
- 在ProteinGym上相关性达0.551，顶级命中率提升367%。
- 适合低资源蛋白工程，减少人工依赖，加速实验验证。
零样本突变预测对低资源蛋白质工程至关重要，但现有蛋白质语言模型（PLMs）常产生看似可信却违背基本生物物理规律的结果。当前候选突变的选择依赖人工专家审核，效率低且主观性强。为此，我们提出Rank-and-Reason（VenusRAR）框架，通过两阶段多智能体协作自动化该流程，最大化湿实验预期适应度。在排序阶段，计算专家与虚拟生物学家构建上下文感知的多模态集成模型，在ProteinGym上实现0.551的斯皮尔曼相关性（优于原有0.518）。在推理阶段，智能体小组采用思维链推理，依据几何与结构约束审计候选突变，在ProteinGym-DMS99上使前5名命中率最高提升367%。对Cas12i3核酸酶的湿实验验证进一步证明其有效性，阳性率达46.7%，并发现两个新突变分别实现4.23倍和5.05倍活性提升。代码与数据集已开源于GitHub。
原文摘要 · Abstract (English)
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/).
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。