This AI ‘Raygun’ can shrink and supersize proteins — opening the door to easy editing
Skip to main content
Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain<br>the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in<br>Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles<br>and JavaScript.
Advertisement
Bluesky
You have full access to this article via your institution.
The protein haemoglobin (artist’s illustration) can be miniaturized by an AI tool called Raygun.Credit: Evan Oto/SPL<br>A sci-fi weapon has entered the AI age. A artificial-intelligence tool dubbed Raygun can miniaturize or supersize a natural protein without disrupting the protein’s shape or ability to function1.<br>Scientist have developed a host of AI ‘protein language models’ that can create proteins from scratch. But Raygun can modify existing proteins, using some of the same steps as natural evolution: adding or deleting single protein subunits or substituting one protein subunit for another.<br>Raygun is a “meaningful and creative” first step towards editing existing proteins, says Fajie Yuan, a computational biologist who works on protein language models at Westlake University in Hangzhou, China, and is not involved in the work. “For many real applications, that is exactly what researchers want — not a completely new protein, but a better, smaller, larger or more adaptable version of one they already trust,” he says.<br>The study is published today in Nature.<br>Protein revolution<br>Protein-building models are now so advanced that they can design antibodies with clinical potential. But scientists would also like to resize natural proteins — a tantalizing prospect for biotechnology applications, says study co-author Rohit Singh, a computational biologist at Duke University in Durham, North Carolina. “Sometimes a smaller protein can do things or fit into places where a larger protein cannot,” he says.<br>But efforts to use AI systems to modify existing proteins have been limited, because they tend to change the structure or function.<br>I told AI to make me a protein. Here’s what it came up with
Singh and his colleagues turned to ESM-2, a large language model for protein design. Instead of text, ESM-2 and other AI-based protein models are trained on millions of protein sequences. After learning the evolutionary ‘grammar’ or patterns of these sequences, these models can generate new proteins.<br>Most protein language models create representations of proteins as amino acid sequences of various lengths. But this system makes it is difficult to create proteins that keep their overall structure and functions at different sizes. So Raygun, which is based on ESM-2, takes a more mathematical approach: it divides each protein into pieces and translates the information in each piece into numerical patterns. The tool then uses these patterns to create a standardized version of the protein that follows certain rules. The model learns those rules, allowing it to generate the protein at various sizes without compromising its structural integrity.<br>Mini-molecules<br>The researchers first used Raygun to shrink four proteins of different sizes, ranging from haemoglobin, which is made up of 147 amino acids, to mTOR, an enzyme that’s composed of 2,549 amino acids. Raygun shaved down the number of amino acids by 15–20% without compromising the protein’s original structure or function. In some cases, it could even halve or double the amino-acid count.<br>Singh and his colleagues also created miniature versions of mCherry and eGFP, fluorescent proteins that are widely used for labelling molecules in biology research. Raygun reduced the number of amino acids in these proteins by 10–16%, making them smaller than nearly all the fluorescent proteins listed in major databases, while maintaining their characteristic shape and ability to glow.<br>‘ChatGPT for CRISPR’ creates new gene-editing tools
At the other end of the scale, Raygun generated a large version of epidermal growth factor (EGF), a protein that kickstarts cell growth and differentiation. In its enlarged form, the protein was able to bind more strongly to its receptor, a common target in cancer therapies. The researchers were surprised by how well Raygun worked. “It was like a shot in the dark,” says study co-author Kapil Devkota, a computational biologist also at Duke University.<br>The results are “encouraging”, but whether Raygun can accomplish other editing tasks is still an open question, says Yuan. It’s also unclear whether the tool would work on other types of proteins, he adds. “We do not yet know whether the same level of control will hold across much broader and more diverse protein families,” says Yuan.<br>Singh and his colleagues are already optimizing Raygun so that it can generate compact and functional proteins with less data....