Google DeepMind 已将其 SynthID 水印技术引入合成生物学领域。这项名为 SynthID Bio 的新工具可以在蛋白质的氨基酸序列或预测的 3D 结构中隐藏一个签名。该实验室于周三在博客文章和《自然》杂志上的一篇论文中公布了这一消息。
Google DeepMind has brought its SynthID watermarking technology to synthetic biology. The new tool, SynthID Bio, hides a signature in the amino acid sequence or predicted 3D structure of a protein. The lab announced it on Wednesday in a blog post and a paper in Nature.
Google DeepMind 表示，通过测试不仅可以找到数字设计中的签名，还能找到物理蛋白质中的签名。在实验室测试中，它没有改变蛋白质的工作方式。
Google DeepMind says a test can find the signature in the physical protein as well as in the digital design. In lab tests, it did not change how the proteins worked.
“对于追踪生物设计的来源而言，SynthID Bio 是拼图中的重要一块，”评估过这项工作的科学政策咨询公司（Science Policy Consulting）生物安全政策专家萨拉·卡特（Sarah Carter）表示。
“SynthID Bio is an important piece of the puzzle for tracking the provenance of biological designs,” said Sarah Carter, a biosecurity policy expert at Science Policy Consulting who reviewed the work.
标记是如何工作的 对于序列，SynthID Bio 会微调氨基酸的选择。对于 3D 结构，它会调整原子坐标。
How the mark works For sequences, SynthID Bio nudges the choice of amino acids. For 3D structures, it adjusts atomic coordinates.
据约翰·蒂默（John Timmer）在 Ars Technica 上报道，该工具运行在 Baker 实验室开发的流行蛋白质设计工具 ProteinMPNN 内部。它使用类似于加密密钥的密钥来建议下一个氨基酸。ProteinMPNN 会拒绝任何不适合正常工作蛋白质的建议。为了检测该标记，软件会使用该密钥扫描整个序列，并测量建议的氨基酸出现的频率。
The tool works inside ProteinMPNN, a popular protein design tool from the Baker Lab, John Timmer reported for Ars Technica. It uses a key, similar to a cryptographic key, to suggest each next amino acid. ProteinMPNN rejects any suggestion that does not fit a working protein. To detect the mark, software scans the whole sequence with the key. It measures how often the suggested amino acids appear.
该团队测试了结合剂——这是一种旨在与其它蛋白质紧密结合的蛋白质。他们使用 AlphaProteo 和带有 SynthID Bio 版本的 ProteinMPNN 设计了这些结合剂。测试涵盖了三个靶标：VEGF-A、SARS-CoV-2 刺突蛋白 RBD 和 PD-L1。带水印的设计在命中率、结合亲和力和序列多样性方面与未加水印的设计相匹配。Adaptyv Bio 为实验室测试提供了帮助。今年 8 月，Anthropic 表示 Claude 已经设计出了可用的结合剂。
The team tested binders, which are proteins built to latch onto other proteins. It designed them with AlphaProteo and a SynthID Bio version of ProteinMPNN. The tests covered three targets: VEGF-A, the SARS-CoV-2 spike protein RBD and PD-L1. The watermarked designs matched unmarked ones on hit rate, binding affinity and sequence diversity. Adaptyv Bio helped with the lab tests. In August, Anthropic said Claude had designed working binders.
对于结构，SynthID Bio 微调了 AlphaFold 3 扩散网络的一小部分。随后，水印便存在于该模型的权重中。Google DeepMind 表示，AlphaFold 3 保持了其准确性，并且检测效果近乎完美。今年 7 月，TNW 报道称 DeepMind 已经解散了 AlphaFold 背后的团队。
For structures, SynthID Bio fine-tunes a small part of AlphaFold 3’s diffusion network. The watermark then sits in the model’s weights. Google DeepMind says AlphaFold 3 keeps its accuracy, and detection is near-perfect. In July, TNW reported that DeepMind had broken up the team behind AlphaFold.
关于DNA订单的生物安全措施：为了合成特定的蛋白质，实验室会从DNA合成公司订购所需的DNA序列。这些公司会将这些订单与已知的生物威胁数据库进行比对，以确保订单的安全性。谷歌DeepMind表示，如今人工智能（AI）已经能够设计出那些在结构上与任何已知有害序列都极为相似的DNA序列；因此，对于这些未知的DNA序列，可能仍需要人工进行仔细审查。为了验证订单的来源可靠性，可以在DNA序列中添加特定的“水印”标记。
A check on DNA orders To make a designed protein, a lab orders DNA from a synthesis company. These companies screen orders against databases of known threats. AI can now design sequences that look little like any known hazard, Google DeepMind says. Unfamiliar orders can then need slow manual reviews. A watermark could show that an order came from a trusted model.
“对于Twist Bioscience公司而言，这种水印技术为生物安全体系带来了新的、非常有用的工具——它能够提升筛查效率，将资源集中在那些需要进一步审查的序列上，并随着AI在生物技术领域的不断发展，使生物安全措施更加高效。”Twist Bioscience公司的政策与生物安全副总裁James Diggans说道。
“For Twist, watermarking offers a promising new addition to the biosecurity toolbox that could strengthen screening, focus resources on sequences that warrant closer review and make biosecurity more efficient as AI-designed biology continues to advance,” said James Diggans, vice president of policy and biosecurity at Twist Bioscience. Google DeepMind says the mark could also help label AI-made entries in public databases.
谷歌DeepMind还表示，这种水印技术还可以用于标记公共数据库中的AI生成的数据（例如Protein Data Bank、UniProt和GenBank等）。由OpenAI支持的Red Queen Bio公司正在利用AI技术来设计针对未来可能出现的病毒的抗体。
It names the Protein Data Bank, UniProt and GenBank. OpenAI-backed Red Queen Bio is using AI to design antibodies for future viruses. Known gaps The team has flagged several limits, according to Ars Technica. The system is only as secure as the process that shares and stores its keys.
存在的局限性：根据Ars Technica的报道，该技术仍存在一些局限性：该系统的安全性取决于用于共享和存储相关数据的流程；非常短的蛋白质序列可能含有太少的可检测标记；将带有标记的蛋白质与未标记的蛋白质混合可能会稀释检测信号；此外，检测结果具有统计性质，因此检测阈值会直接影响假阳性和假阴性的发生率。
Very short proteins may carry too few marked amino acids to detect. Fusing a marked protein with an unmarked one could dilute the signal. Detection is statistical, so the cut-off sets the rate of false positives and false negatives.
未来的发展方向：许多AI蛋白质设计工具并未采用Google DeepMind提出的这种“ProteinMPNN”技术；该公司表示仍需进一步改进该技术，以使其标记更加难以被人为删除。
Many AI protein design tools do not use ProteinMPNN. Google DeepMind says it still needs to make the mark harder to remove on purpose.
针对细菌的病毒研究：谷歌DeepMind还与斯坦福大学的Hie实验室及Arc研究所合作，将“SynthID Bio”技术整合到了名为Evo 2的基因组模型中。他们共同对Evo 2生成的噬菌体基因组添加了水印标记。实验表明，这种带有水印的噬菌体在细菌培养环境中能够正常发挥作用。该公司计划发表相关的技术论文。
Next, viruses that infect bacteria Google DeepMind has also added SynthID Bio to Evo 2, a genomic model, with the Hie lab at Stanford University and Arc Institute. Together they watermarked the genome of a bacteriophage that Evo 2 designed. A bacteriophage is a virus that infects bacteria. Early tests in bacteria cultures show the watermarked phages work, the lab says. It plans to publish a technical paper.
谷歌DeepMind已将相关代码和实验数据开源，并向研究人员提供了模型参数（即模型的权重数据）。显微镜争议中的SynthID据Alex Blake为TechRadar报道，谷歌用于图像和视频的SynthID标记也处于另一场争议的中心。尼康正在重新审查其“微观世界动态”（Small World in Motion）比赛的获奖视频。该视频由清华大学的徐宁博士拍摄，展示了一个患有罕见遗传病的儿童的气道纤毛。视频的说明文字写着“后期制作中使用了AI辅助”。
Google DeepMind has made the code and lab data open source. It is also releasing the model weights to researchers. SynthID in a microscopy dispute Google’s SynthID mark for images and video is also at the centre of a separate dispute. Nikon is re-reviewing the winning video in its Small World in Motion contest, Alex Blake reported for TechRadar. The video, by Dr Ning Xu of Tsinghua University, shows airway cilia from a child with a rare genetic disorder. Its caption reads “AI-assisted in post-processing”.
据TechRadar报道，德克萨斯大学西南医学中心的博士生伊恩·多诺万（Ian Donovan）表示，他在该视频中发现了嵌入的SynthID水印。尼康在领英（LinkedIn）上表示，徐宁已经提供了详细的技术文档。
Ian Donovan, a PhD student at UT Southwestern Medical Center, said he had found an embedded SynthID watermark in the video, according to TechRadar. Nikon said on LinkedIn that Xu had provided detailed technical documentation.
据TechRadar报道，徐宁在领英上表示：“人工智能并未被用于生成实验电影、纤毛或它们的运动。”
“AI was not used to generate the experimental movie, the cilia, or their motion,” Xu said on LinkedIn, according to TechRadar.
徐宁表示，他曾使用人工智能来辨别和可视化重建灰度图像中的特征。
Xu said he had used AI to distinguish and visualise features in the reconstructed greyscale images.