2026年10月16日 (英语).
Pooja Rao，研究科学家，软件工程师Petar Sirkovic和研究科学家Annalisa Pawlosky,Google
我们把AI agents带到Google研究的湿板凳上，以导航复杂的生物景观。 通过将agents放置在物理实验室机器人中，我们可以运行快速的闭路分子设计周期，agents学习如何学习，最终发现单位数纳米蛋白结合物，用于癌靶。
边疆AI系统有可能通过智能导航广阔的组合搜索空间(如单个100氨基酸蛋白至少跨20100个可能的序列)来改变生物学研究和药物发现，要求动态的多扫描探索。 我们的长期研究愿景是AI agents不仅会进行想象和分析，而且还会进行实验室实验来测试他们的假说，学习结果，以及自我改进来解决复杂的问题——这一切都是由人类科学机构推动的。
实现这一愿景需要解决两个相互交织的挑战：第一，前沿模型仍然无法可靠地解释三维几何和物理状态——这给专门液体处理机器人自动化带来了问题，这些机器人可以移动试剂，跟踪体积，并在湿实验室执行分子生物学协议。 第二，虽然代理假说生成和内硅化设计正在迅速推进，但需要反复循环的实时世界反馈来验证提案并导航广阔的搜索空间，但现实世界的湿板实验并没有跟上步伐，使实验读取比以往更加成为瓶颈。
本文分享了三个研究项目
- 使AI agents能够利用核查数据数据库框架，在物理测量的基础上进行安全、资源高效的实验。
- 展示我们原型上的闭路剂科学发现用于patamer和去新蛋白设计
首先，我们提出 " 经核实的发现实验室基础设施 " 框架，以弥合AI agents高级别推理与湿实验室复杂、容易出错的物理现实之间的差距。 它是实验室工作细胞的语义数字双胞胎，即活的，状态的3D体积实体树，跟踪从机器人臂几何到单个井体体积，试剂身份，以及管状尖端盘点等一切情况。
命令的安全性取决于现场物理环境以及代码。 agent可以生成物理上不可能的合成有效的代码(例如将一个8个通道的管子降低到一个太短的狭窄槽中)，后勤上无效(例如从耗尽的井中呼吸)，或者受到未建模的流体效应的科学损害。
建于开源 PyLabrobot框架之上，VERID提供了四个核心能力：
- Live 3D状态跟踪：允许agents在行动前检查甲板布局，音量和可用的提示。
- 快速虚拟干燥运行：在触碰试剂前，以秒模拟协议以捕捉碰撞或空井。
- 硬件安全检查：在到达机器人之前，封锁无效的命令。
- 同步视频和动作日志：收集每一次运行的完整记录，用于研究精度、可复制性和调试。
在我们的报告中，我们通过个案研究显示这些行动能力。 首先，同步的视频和处决痕迹被捕获的染料分层不均匀，因为射出离井层太近，使得agent能够完善其混合策略。 在同一场战役中，持续的数字双子状态使得agent能够从板块阅读器故障中恢复中场，而无需重排。 第二，当模拟器拒绝一个实际无法融入狭小槽的8通道的欲望时，agent在苍蝇上重新规划，并设计了定向的因子诊断隔离因残余的尖墙传承而导致的微妙偏差，单是复制就会错过。 最后，agent在一次闭路自玩活动中，迭代设计并运行了机器人实验，从俯冲相机影像中重建失明染料混合物，识别自己的光学和流体文物横跨，提高精度超过5倍。
可复制的执行和完全的追踪日志——捕捉像居住时间，蒸发，板块布局等程序变量——为agents在探索和精炼许多候选者时最终能够将生物信号与实验噪声区分开来奠定了基础。
Agents需要真正的实验读取来完善他们的假说，因为计算模型不能捕捉湿板反应中发生的事情的全貌，不能取代经验验证。 在“Aptamers as a sandbox for lab-in-the-loop agency 分子发现”中，我们使用蛋白质结合DNA的aptamers在3个设计-测试-渗入周期上与AI agents进行密闭循环对实验室进行原型。 阿普塔默发现(英語：Aptamer discoveration)是临床生物标志器发现研究努力中常见的一种实验。 一个完整的aptamer设计-测试-分析周期只需要7-10天，因为DNA可以在集合库中化学合成，并通过测序直接读出。 在每个回合中，agents设计候选序列，这些序列在磁珠上与目标蛋白进行筛选，以隔离功能绑定器，然后进行排序，以返回下一个周期丰富的定量数据。
在两组不同难度的蛋白质靶点(迷你红蛋白和α-色素)上，AI驱动的理性和新设计融资确认了目标命中率，如下图所示(肌红蛋白和α-色素分别为33.5%和42.1%)，尽管每个靶点都需要识别和克服不同的湿板故障模式。
在α色素上，第1轮产生0%的粘合剂，因为固定的底架中断了48%的序列的折叠；一旦agents 推出夹住的侧翼出现在第2轮，并在第3轮多样化，命中率上升至26.0%，然后以98.5%的特异性达到42.1%。
在 myoglobin上，由于agents在第2轮中优先考虑原始的绑定亲和性，它们高度浓缩的G-quadruplex motifs(占设计的63%). 虽然原始的拉倒增加，但湿盘数据显示，这些G-quadruplex大多是非特定的"粘滞"粘附剂，坚持非目标蛋白和裸磁珠，导致目标特异性(从75.0%到19.1%)和真实的，经珠校正的命中率(从10.2%到3.7%)都暴跌。 agents认识到第2轮读取的这一文物，自动修订了第3轮的设计规则，将断面齿轮封顶，减去珠状粘合器，向非G4结构脚手架支引，以恢复特异性(72.4%)，并在目标命中率上猛增9倍(33.5%)。
在论文中，我们还研究了agents如何像科学家那样的理性，分析数据，向湿实验室协议建议微调，但有时也会非典型的思考，探索结构生物学家可能不会的序列空间。
关闭蛋白质循环更为复杂。 与化学合成的aptamers不同，蛋白质必须用生物表达，因此每个循环新增细胞或无细胞选择，表达，和折叠步骤。 加上较大的序列空间，这在计算和湿板两侧都造成了瓶颈。
在“Agent制导的直肠切除法能够实现经济的重新绑定剂设计”中，我们把工作流程应用于纤维素激活蛋白(FAP)，这是一种在固体肿瘤中过度表达的具有挑战性的大型受体。 仅在少数实验中，顶部设计产生了带有单数纳米摩尔亲和度的绑定剂，这种紧凑，类似药物的绑定度一般只有在经过大量几轮实验室优化后才能实现。
计算，de novo 绑定器设计需要生成候选骨干，设计序列以适应它们，并重新叠加由此产生的复合体以预测绑定。 最后的重叠步骤消耗了绝大多数GPU计算，因为它与目标长度成四倍的缩放。 我们的输油管在折叠过程中通过手术排泄出局部的束缚口袋，加快了每个循环的8倍以上，而不失去结构精度。 这转而决定在何处捆绑，将多少周围结构纳入LLM agents非常适合优化的战术杠杆。
在同等计算法的正面比较中，人类-agent合作，即科学家设定高级策略，agent驱动战术执行，几乎是科学家单独工作的西里科命中率的两倍，同时在更广泛的目标地点发现达到。 为了在计算管道的同时扩大湿盘筛选，我们用荧光激活的细胞分类(FACS)操作酵母表面显示，这个工作流程使我们能够通过荧光同步筛选工程蛋白变体的大型库。
我们的下一步是把这些线条编织在一起，把VERID的完全可追踪的自动化和闭路学习用于蛋白质发现，使agents能够直接导航湿实验室中复杂而崎岖的生物搜索空间。
根据Google的《边境安全框架》和我们处理生物复原力问题的方法，我们在工作流程的计算和湿板两侧建立保障措施：
- 安全试验床和检查：我们有意选择控制良好、没有感染力的生物系统，而不产生功能风险，作为评价agent行为和及早捕捉漂流的试验床。 每个设计序列都要经过生物安保筛选。
- 执行前的模拟：通过VERID,agents从不直接向液体处理机器人发送未经验证的代码。 每个协议在模拟中都是干燥的，在触摸试剂之前对数字双体进行验证。
- 人类监督：人类直接授权和审查设计的序列，核查湿实验室协议，并监督机器人运行。
随着AI agents承担更多的设计-测试周期，出现了新的安全和可靠性挑战。 走向一个自主的湿实验室，需要用人类代理和监督的机会来解决这些开放的问题：确保agents优化真实的目标绑定而不是意外的实验快捷方式，验证多步骤的湿实验室协议，并持续监测agent推理和执行痕迹。
从agents在物理实验室机器人中搁浅，到运行快速闭合-loop aptamer循环，其中agents学习探索序列空间，到发现单数纳米莫尔蛋白结合物，这些努力构成了生命科学中人类定向自主发现的构件。 我们对这一进展感到兴奋，并期待着分享接下来的情况。
这项研究由安娜利莎·帕洛斯基(Annalisa Pawlosky)领导，与波雅·拉奥(Pooja Rao)和佩塔尔·西尔科维奇(Petar Sirkovic)并列。 我们要感谢支持这项工作的基本文件的共同作者和合作者：Zach Cutts、John Coller、Rachel Chen、Mathias Bellaiche、Hua Tang、Ryder Easterlin和Dina Zverinski(aptamer发现);Jun Kim、Artiom Myaskovsky、Oscar Donosa、Jamie Rogers和Lawrence Chan(蛋白质设计)；以及Yves Delley、John Coller、Zach Cutts、Alexander Daryin、Mathias Bellaiche、Jun Kim、Oscar Donosa、Srinidhi Seshadri、Tanner Clary和Sandy Klemm，另外还有来自辛伊·登、Sanath Ramacandran和Jiawu Bi(VRID自动化)。
我们深为感谢雷文·萨亚格、阿维纳坦·哈西迪姆、罗尼特·莱瓦维·莫拉德、约西·马蒂亚斯和詹姆斯·马尼卡的领导指导，并感谢马蒂亚斯·海勒、布拉克·戈克图尔克、比尔·贾和普希米特·科赫利在整个研究期间的支持。 我们还感谢谢科菲·阿齐齐、安德鲁·卡罗尔、菲利普·纳尔逊、伊沙伊·曼苏尔、丹·波波维奇和里卡多·皮索尼的回顾和富有洞察力的反馈、阿马尼·布里克对这项努力的全面贡献以及乌迪·格拉夫的视觉。
本文由 AI宝盒自动抓取并翻译，版权归原作者所有；如需查看原文请点下方链接。
October 16, 2026
Pooja Rao, Research Scientist, Petar Sirkovic, Software Engineer, and Annalisa Pawlosky, Research Scientist, Google
We are bringing AI agents to the wet-lab bench at Google Research to navigate complex biological landscapes. By grounding agents in physical lab robotics, we can run fast closed-loop molecular design cycles where agents learn how to learn, ultimately discovering single-digit nanomolar protein binders for cancer targets.
Frontier AI systems have the potential to transform biology research and drug discovery by intelligently navigating vast combinatorial search spaces (e.g., a single 100-amino-acid protein spans at least 20100 possible sequences) calling for dynamic multi-candidate exploration. Our long-term research vision is that AI agents will not only ideate and analyze, but also run lab experiments to test their hypotheses, learn from the results, and self-improve to solve complex problems — all driven by human scientific agency.
Realizing this vision requires solving two intertwined challenges: First, frontier models still do not reason reliably about 3-dimensional geometry and physical state — which poses a problem in the automation of the specialized liquid-handling robots that move reagents, track volumes, and execute molecular biology protocols in a wet lab. Second, while agentic hypothesis generation and in-silico design are advancing rapidly, needing repeated cycles of real-world feedback to validate the proposals and navigate vast search spaces, real-world wet-lab experiments have not kept pace, making experimental readouts more of a bottleneck than ever.
In this post we share three research projects highlighting our progress on bridging the gap between in-silico AI exploration and real-world wet-lab experimentation:
- Enabling AI agents to perform safe, resource efficient experimentation grounded in physical measurements by using the VERID framework.
- Demonstrating closed-loop agentic scientific discoveries on our prototypes for aptamer and de-novo protein design.
First, we present the VERified Infrastructure for Discovery Laboratories (VERID) framework to bridge the gap between the high-level reasoning of AI agents and the complex, error-prone physical reality of a wet lab. It is a semantic digital twin of the laboratory workcell, i.e., a live, stateful 3D volumetric entity tree that tracks everything from robotic arm geometry to individual well volumes, reagent identities, and pipette tip inventory.
The safety of a command depends on the live physical context as well as the code. An agent can generate syntactically valid code that is physically impossible (e.g., lowering an 8-channel pipette into a narrow trough that is too short), logistically invalid (e.g., aspirating from a depleted well), or scientifically compromised by unmodeled fluidic effects.
Built on top of the open-source PyLabRobot framework, VERID provides four core capabilities:
- Live 3D state tracking: Allow agents to check deck layout, well volumes, and available tips before acting.
- Fast virtual dry runs: Simulate protocols in seconds to catch collisions or empty wells before touching reagents.
- Hardware safety checks: Block invalid commands before they reach the robot.
- Synchronized video and action logs: Capture a complete record of every run for research rigor, reproducibility, and debugging.
In our report, we show these capabilities in action through case studies. First, synchronized video and execution traces caught uneven dye stratification caused by dispensing too close to the well floor, allowing the agent to refine its mixing strategy. In the same campaign, the persistent digital twin state allowed the agent to recover mid-run from a plate-reader failure without re-pipetting. Second, when the simulator rejected an 8-channel aspiration that physically wouldn't fit in a narrow trough, the agent replanned on the fly and designed a targeted factorial diagnostic isolating a subtle bias caused by residual tip-wall carryover that replication alone would have missed. Finally, in a closed-loop self-play campaign, an agent iteratively designed and ran robot experiments to reconstruct blinded dye mixtures from overhead camera images, identifying its own optical and fluidic artifacts across runs to improve accuracy more than fivefold.
Reproducible execution and full trace logs — capturing procedural variables like dwell times, evaporation, and plate layout — lay the groundwork for agents to eventually be able to separate biological signals from experimental noise as they explore and refine many candidates in parallel across rounds.
Agents need real experimental readouts to refine their hypotheses as computational models do not capture the full picture of what happens in a wet-lab reaction and cannot replace empirical validation. In “Aptamers as a sandbox for lab-in-the-loop agentic molecular discovery”, we use protein-binding DNA aptamers to prototype the lab in a closed loop with AI agents over 3 design-test-iteration cycles. Aptamer discovery is a type of experiment common in research efforts for clinical biomarker discovery. A complete aptamer design-test-analyze cycle takes just 7–10 days, as DNA can be chemically synthesized in pooled libraries and read out directly by sequencing. In each round, agents design candidate sequences, which are screened against target proteins on magnetic beads to isolate functional binders, and then sequenced to return rich quantitative data for the next cycle.
Across three rounds on two protein targets with varying difficulty levels (myoglobin and α-thrombin), AI-driven rational and de novo design raised confirmed on-target hit rates as can be seen in the image below (to 33.5% and 42.1% for myoglobin and α-thrombin, respectively), though each target required identifying and overcoming a different wet-lab failure mode.
On α-thrombin, Round 1 yielded 0% binders because fixed primer overhangs disrupted folding in 48% of sequences; once the agents introduced clamped flanking stems in Round 2 and diversified in Round 3, hit rates climbed to 26.0% and then 42.1% at 98.5% specificity.
On myoglobin, as agents prioritized raw binding affinity in Round 2, they heavily enriched G-quadruplex motifs (63% of designs). While raw pull-down increased, the wet-lab data showed that most of these G-quadruplexes were non-specific "sticky" binders adhering to off-target proteins and bare magnetic beads, causing both target specificity (from 75.0% to 19.1%) and true, bead-corrected hit rate (from 10.2% to 3.7%) to plunge. Recognizing this artifact from the Round 2 readout, the agents autonomously revised their design rules for Round 3, capping promiscuous motifs, subtracting bead binders, and pivoting to non-G4 structural scaffolds to recover specificity (72.4%) and surge ninefold in on-target hit rate (33.5%).
In the paper, we also examine how agents reason like scientists, analyzing data and suggesting tweaks to wet-lab protocols, but also sometimes think atypically, exploring sequence spaces that structural biologists might not.
Closing the loop for proteins is more complex. Unlike aptamers, which are chemically synthesized, proteins must be expressed biologically, so each loop adds cellular or cell-free selection, expression, and folding steps. Together with a larger sequence space, this creates bottlenecks on both the computational and wet-lab sides.
In “Agent-guided epitope excision enables economical de novo binder design”, we applied our workflow to fibroblast activation protein alpha (FAP), a large, challenging receptor overexpressed in solid tumors. In just a few experimental runs, the top designs yielded binders with single-digit nanomolar affinity, a level of tight, drug-like binding typically achieved only after extensive rounds of laboratory optimization.
Computationally, de novo binder design requires generating candidate backbones, designing sequences to fit them, and re-folding the resulting complexes to predict binding. That final re-folding step consumes the vast majority of GPU compute, as it scales quadratically with target length. By surgically excising just the local binding pocket during folding, our pipeline accelerated each cycle more than eightfold without losing structural accuracy. This turns decisions about where to bind and how much surrounding structure to include into tactical levers that LLM agents are well-suited to optimize.
In a head-to-head comparison at equal compute, human–agent collaboration, where a scientist sets the high-level strategy and an agent drives tactical execution, nearly doubled the in-silico hit rate of a scientist working alone while uncovering hits across a much broader range of target sites. To scale wet-lab screening alongside our computational pipeline, we operationalized yeast surface display with fluorescence-activated cell sorting (FACS), a workflow that allows us to screen large libraries of engineered protein variants in parallel by fluorescence.
Our next step is to weave these threads together, combining VERID’s fully traceable automation with closed-loop learning for protein discovery to enable agents to directly navigate complex, rugged biological search spaces in the wet lab.
In line with Google’s Frontier Safety Framework and our approach to bioresilience, we build safeguards across both the computational and wet-lab sides of our workflows:
- Safe testbeds and screening: We deliberately selected well-controlled, non-infectious biological systems without gain-of-function risk, as testbeds to evaluate agent behavior and catch drift early. Every designed sequence goes through biosecurity screening.
- Simulation before execution: Through VERID, agents never send unverified code directly to the liquid handling robots. Every protocol is dry-run in simulation and validated against the digital twin before touching reagents.
- Human oversight: Humans directly authorize and review designed sequences, verify wet-lab protocols, and supervise robot runs.
As AI agents take on more of the design-test cycle, new safety and reliability challenges emerge. Moving toward an autonomous wet lab requires solving these open problems with opportunities for human agency and oversight: ensuring agents optimize for true target binding rather than unintended experimental shortcuts, verifying multi-step wet-lab protocols, and continuously monitoring agent reasoning and execution traces.
From grounding agents in physical lab robotics, to running fast closed-loop aptamer cycles where agents learn to explore sequence space, to discovering single-digit nanomolar protein binders, these efforts form the building blocks for human-directed autonomous discovery in the life sciences. We're excited by this progress and look forward to sharing what comes next.
This research is led by Annalisa Pawlosky, alongside Pooja Rao and Petar Sirkovic. We would like to thank our co-authors and collaborators across the foundational papers supporting this work: Zach Cutts, John Coller, Rachel Chen, Mathias Bellaiche, Hua Tang, Ryder Easterlin, and Dina Zverinski (aptamer discovery); Jun Kim, Artiom Myaskovsky, Oscar Donosa, Jamie Rogers, and Lawrence Chan (protein design); and Yves Delley, John Coller, Zach Cutts, Alexander Daryin, Mathias Bellaiche, Jun Kim, Oscar Donosa, Srinidhi Seshadri, Tanner Clary, and Sandy Klemm, with additional technical and experimental contributions from Xinyi Deng, Sanath Ramachandran and Jiawu Bi (VERID automation).
We are deeply grateful to Reuven Sayag, Avinatan Hassidim, Ronit Levavi Morad, Yossi Matias, and James Manyika for their leadership guidance and to Matthias Heiler, Burak Gokturk, Bill Jia and Pushmeet Kohli for their support throughout this research. We also thank Shekoofeh Azizi, Andrew Carroll, Philip Q. Nelson, Yishay Mansour, Dan Popovici, and Riccardo Pisoni for their reviews and insightful feedback, Amanie Brik for overall contributions to the effort, and Udi Graff for visual des
Auto-fetched and translated by AI Gem Box. All rights belong to the original author.