Anthropic Says It Discovered a Crispr-Like System. Now What?
Anthropic Says It Discovered a Crispr-Like System. Now What?
Anthropic 称其发现了一种类似 CRISPR 的系统，接下来会怎样？
Tech executives have long touted the potential for their AI models to accelerate the pace of biology discoveries. Last week, they announced one. In a September 23 announcement, Anthropic claimed that its large language model Claude had identified an enzyme system with properties “reminiscent of Crispr,” the Nobel Prize-winning gene-editing tool.
科技高管们长期以来一直吹捧其人工智能模型在加速生物学发现方面的潜力。上周，他们宣布了一项成果。在 9 月 23 日的一份公告中，Anthropic 声称其大语言模型 Claude 发现了一种酶系统，其特性“让人联想到 CRISPR”——即那项获得诺贝尔奖的基因编辑工具。
According to Anthropic, about 950 Claude agents running simultaneously found the interesting genetic sequences in 21.5 hours. “I sincerely compliment Anthropic for telling the world about their discovery,” says Fyodor Urnov, a gene-editing expert at the University of California, Berkeley and director for therapeutic R&D at its Innovative Genomics Institute. (IGI is collaborating with Anthropic, but was not involved with the new discovery.)
据 Anthropic 称，约 950 个 Claude 智能体同时运行，在 21.5 小时内找到了这些有趣的基因序列。“我真诚地称赞 Anthropic 向世界公布了他们的发现，”加州大学伯克利分校基因编辑专家、创新基因组学研究所（IGI）治疗研发主任 Fyodor Urnov 表示。（IGI 正在与 Anthropic 合作，但并未参与此次新发现。）
Other scientists are less enthusiastic about the finding, which is the first to come out of a research group formed by Anthropic earlier this year. “The experiments are still in the queue. The PR is already live,” says Le Cong, a professor at Stanford University focused on integrating AI into genome engineering research.
其他科学家对这一发现的热情则较低，这是 Anthropic 今年早些时候成立的研究小组发布的首个成果。“实验还在排队中，公关稿却已经发布了，”斯坦福大学教授 Le Cong 说道，他专注于将人工智能整合到基因组工程研究中。
Anthropic, which has set up a wet lab for drug discovery, has been clear that the finding isn’t the end of its work. (“We don’t yet understand what this system does,” the company said in an X post, “but only a handful of known systems share its features, and all of them are able to cut, copy, and paste DNA.”) One physical experiment was included in a technical report, which has not been peer-reviewed. However, a lot more needs to be done to determine the significance of this discovery. It remains unknown whether this enzyme system can be used as a gene-editing tool and, if so, whether it’s a useful gene-editing tool.
Anthropic 已经建立了一个用于药物发现的湿实验室，并明确表示这一发现并非其工作的终点。（该公司在 X 上发帖称：“我们还不了解这个系统具体的功能，但只有少数已知系统具备其特征，且它们都能切割、复制和粘贴 DNA。”）一份技术报告中包含了一项物理实验，但该报告尚未经过同行评审。然而，要确定这一发现的意义，还需要做更多的工作。目前尚不清楚该酶系统是否可以用作基因编辑工具，如果可以，它是否是一种有用的基因编辑工具。
“Let’s say we are on Santa Monica Beach and trying to scan through all the sand to find a diamond,” Cong says. “AI found this thing that looks very shiny, and then you have to go back to the lab to know—is it glass? Is it a diamond?”
“假设我们在圣莫尼卡海滩，试图扫描所有的沙子来寻找一颗钻石，”Cong 说，“人工智能找到了看起来闪闪发光的东西，然后你必须回到实验室去确认——它是玻璃？还是钻石？”
Claude found this shiny object after researchers at Anthropic prompted it to search through huge genomic databases for “interesting new examples” of reverse transcriptases, proteins that copy RNA into DNA. This is the opposite of the normal process in our cells, in which DNA is transcribed into RNA, but is used by other organisms in a variety of biological functions, including inserting segments of DNA.
Claude 在 Anthropic 的研究人员提示它在庞大的基因组数据库中搜索逆转录酶（将 RNA 复制为 DNA 的蛋白质）的“有趣新样本”后，发现了这个“闪光物体”。这与我们细胞中的正常过程（DNA 转录为 RNA）相反，但被其他生物用于多种生物功能，包括插入 DNA 片段。
Claude agents initially identified more than 200,000 possible reverse transcriptases, according to Anthropic, and, after picking out several thousand that appeared to be new, narrowed the scope even further, eventually finding an “unusual” reverse transcriptase family containing a long region of repeat DNA sequences that resemble Crispr. Anthropic is calling this system ART, short for array-associated reverse transcriptases. The system is found in jumbo phages, large viruses that infect bacteria.
据 Anthropic 称，Claude 智能体最初识别出超过 20 万种可能的逆转录酶，在筛选出数千种看起来是新的样本后，进一步缩小了范围，最终发现了一个“不寻常”的逆转录酶家族，其中包含一段类似于 CRISPR 的长重复 DNA 序列区域。Anthropic 将该系统称为 ART（阵列相关逆转录酶的缩写）。该系统存在于巨型噬菌体中，这是一种感染细菌的大型病毒。
“I can see by eye a tandem repeat array … that’s a Crispr-like … repeat array?!” an AI agent wrote. The agent also acknowledged that this system could be a retron, as noted in the technical report. Crispr and retrons are both bacterial immune systems, and while retrons have some utility in gene editing, they’re not the same multi-tool Crispr is. It’s therefore not too surprising Anthropic’s blog post played up the Crispr comparison.
“我肉眼就能看到一个串联重复阵列……那是一个类似 CRISPR 的……重复阵列？！”一个人工智能智能体写道。该智能体还承认，正如技术报告中所述，该系统可能是一种逆转子（retron）。CRISPR 和逆转子都是细菌免疫系统，虽然逆转子在基因编辑中有一定用途，但它们并不像 CRISPR 那样是多功能工具。因此，Anthropic 的博客文章大肆渲染与 CRISPR 的对比也就不足为奇了。
Seth Shipman, associate investigator at the Gladstone Institutes, doesn’t think what Anthropic found is a new Crispr system, but says the discovery is still interesting. “The novel thing is how they found it, not what it is,” he says. His lab has used retrons to make gene-editing systems, so it is possible to use retrons in this way.
格拉德斯通研究所副研究员 Seth Shipman 认为 Anthropic 发现的并不是一种新的 CRISPR 系统，但他表示这一发现仍然很有趣。“新颖之处在于他们是如何发现它的，而不是它本身是什么，”他说。他的实验室曾利用逆转子制造基因编辑系统，因此以这种方式使用逆转子是可能的。
Identifying new reverse transcriptases can take months of mining genome databases manually, so Shipman says the fact that Anthropic was able to do this in a day is impressive. But he cautions against attributing the discovery entirely to an AI model. “I think we have to be careful about saying that Claude autonomously discovered something, because there are scientists involved in the study,” he says.
手动挖掘基因组数据库以识别新的逆转录酶可能需要数月时间，因此 Shipman 表示，Anthropic 能在一天内完成这项工作令人印象深刻。但他提醒不要将这一发现完全归功于人工智能模型。“我认为我们在说 Claude 自主发现了什么时必须谨慎，因为研究中有科学家参与，”他说。
In fact, this same reverse transcriptase was previously identified by Jason Gill, a microbiologist at Texas A&M University, and his colleagues in a 2021 paper on jumbo phages. What’s perhaps new is that Claude identified repeats around it that hinted the enzyme might be part of a Crispr-like system.
事实上，这种逆转录酶此前已被德克萨斯农工大学的微生物学家 Jason Gill 及其同事在 2021 年关于巨型噬菌体的论文中识别出来。可能的新发现是，Claude 识别出了其周围的重复序列，暗示该酶可能是类似 CRISPR 系统的一部分。
“These models are good at finding patterns, better than a person staring at it with their eyeballs can,” Gill says. “A person could do all this but you have to already have a hypothesis in mind.” He adds that ART appears to have no “obvious relationship” to any known Crispr system, and it’s on Anthropic to prove that it actually has gene-editing activity.
“这些模型擅长寻找模式，比人眼盯着看要强，”Gill 说，“一个人也可以做所有这些工作，但你必须已经有了假设。”他补充说，ART 似乎与任何已知的 CRISPR 系统没有“明显关系”，Anthropic 有责任证明它确实具有基因编辑活性。
Cong credits the scientists for finding the perfect kind of problem for Claude to tackle, given its pattern-seeking abilities. It’s possible the model could be used for similar large-data projects, but that doesn’t mean Claude will suddenly be accelerating all sorts of discoveries in the life sciences.
Cong 称赞科学家们为 Claude 找到了完美的攻克课题，考虑到其寻找模式的能力。该模型有可能被用于类似的大数据项目，但这并不意味着 Claude 会突然加速生命科学领域的所有发现。
What kinds of information this model was trained on is also not something the scientific community can evaluate, which makes these findings unusually opaque in an era where major journals require scientists to freely publish their code in the name of reproducibility. This has also led a researcher who had been studying these exact enzymes to become suspicious that the model was trained on his work, since he’d shared yet unpublished findings with the public version of Claude. (“Claude was,” Anthropic told the New York Times, “not trained on any user transcripts, and our molecular biology team has no such access, either.”)
该模型是在什么样的数据上训练的，科学界也无法评估，这使得这些发现在当今主要期刊要求科学家以可重复性为名公开代码的时代显得异常不透明。这也导致一位一直在研究这些特定酶的研究人员怀疑该模型是否使用了他的工作成果进行训练，因为他曾与公开版本的 Claude 分享过尚未发表的研究结果。（Anthropic 对《纽约时报》表示：“Claude 没有接受过任何用户对话记录的训练，我们的分子生物学团队也没有此类访问权限。”）
Even if Claude did identify the next big thing in gene editing, it would still take human scientists testing out that hypothesis in the lab to prove it. In a post on X about the discovery, Anthropic CEO Dario Amodei mused, “Eventually it may even be possible for Claude itself to safely perform the experiments by autonomously controlling lab equipment.”
即使 Claude 确实发现了基因编辑领域的下一个重大突破，仍需要人类科学家在实验室中验证该假设才能证明其有效性。在关于这一发现的 X 帖子中，Anthropic 首席执行官 Dario Amodei 沉思道：“最终，Claude 本身甚至可能通过自主控制实验室设备来安全地进行实验。”