{"schema":"hottub-news/story/v1","story":"st_23bcnm6qjldj3ckkmxja","title":"生物安全领域里程碑：谷歌 DeepMind 首创 AI 蛋白质水印技术","title_en":"Milestone in biosafety: Google DeepMind pioneers AI protein watermark technology","outlets":1,"brief":null,"page":"https://hottub.news/stories/milestone-in-biosafety-google-deepmind-pioneers-ai-protein-watermark-23bcnm6qjl","count":1,"items":[{"id":"n_23bcnm6qjldj3ckkmxja","seq":1035096,"kind":"article","title":"生物安全领域里程碑：谷歌 DeepMind 首创 AI 蛋白质水印技术","title_en":"Milestone in biosafety: Google DeepMind pioneers AI protein watermark technology","url":"https://www.ithome.com/1/008/940.htm","summary":"IT之家 9 月 30 日消息，谷歌今日推出 SynthID Bio ，将数字水印技术引入合成生物学领域。该技术可将不可感知的签名直接嵌入生物编码，不仅能在数字模型上验证， 还可在合成的物理蛋白质上完成验证，同时在实验室测试中保持蛋白质的生物学功能。 生成式 AI 正在帮助科学家解决关键生物学挑战，从预测蛋白质结构到设计全新蛋白质，再到开发新型噬菌体。但这类工具也带来了新挑战：新型 AI 设计可能绕过传统 DNA 合成筛选， 而标记错误的合成 3D 结构会污染公共数据库，误导后续研究。 SynthID Bio 是专门为合成生物学开发的水印方法系列，旨在加强生物安全和科学诚信。 它会根据数据类型调整方法，对序列的氨基酸选择进行 subtle 引导，对预测的 3D 结构调整原子坐标，从而创建可靠的检测信号。实验表明，这些调整不会损害蛋白质的生物学功能，这对于有效治疗疾病和推进科学研究至关重要。 在针对 VEGF-A、SARS-CoV-2 刺突蛋白 RBD 和 PD-L1…","published":"2026-10-01T01:28:53Z","seen":"2026-10-01T02:15:12Z","lang":"zh","country":"CN","source":{"id":"kite-ithome-com-f1cdd0","name":"ithome.com","domain":"ithome.com","outlet":{"wikidata":"Q28413779","name":"IThome","type":"website","country":"CN","designations":["website"]}},"via":"kite-ithome-com-f1cdd0","entities":[{"id":"Q15733006","name":"Google DeepMind","type":"org"}],"story":"st_23bcnm6qjldj3ckkmxja","category":"science-technology","category_p":0.9700000286102295,"sentiment":"positive","tone":0.46000000834465027,"political":0.05000000074505806}]}