{"id":1997,"date":"2026-04-09T21:17:08","date_gmt":"2026-04-09T21:17:08","guid":{"rendered":"https:\/\/www.europesays.com\/ai\/1997\/"},"modified":"2026-04-09T21:17:08","modified_gmt":"2026-04-09T21:17:08","slug":"ai-diffusion-models-tailor-drug-molecules-to-custom-fit-protein-targets-speeding-drug-development-and-evaluation","status":"publish","type":"post","link":"https:\/\/www.europesays.com\/ai\/1997\/","title":{"rendered":"AI diffusion models tailor drug molecules to custom-fit protein targets, speeding drug development and evaluation"},"content":{"rendered":"

\"New<\/p>\n

Structural and topological representation of protein\u2013compound interactions in YuelPocket’s graph neural network framework. Credit: Proceedings of the National Academy of Sciences (2026). DOI: 10.1073\/pnas.2524913123<\/p>\n

University of Virginia School of Medicine scientists have developed a bold new approach to drug development and discovery that could dramatically accelerate the creation of new medicines. UVA’s Nikolay V. Dokholyan, Ph.D., and colleagues have developed a suite of artificial intelligence-powered tools, called YuelDesign, YuelPocket and YuelBond, that work together to transform how new drugs are created. The centerpiece, YuelDesign, uses a cutting-edge form of AI called diffusion models to design new drug molecules tailored to fit their protein targets exactly, even accounting for the way proteins flex and shift shape during binding.<\/p>\n

A companion tool, YuelPocket, identifies exactly where on a protein a drug can attach, while YuelBond ensures the chemical bonds in designed molecules are accurate. Together, the approach is poised to improve both how new drugs are designed and how quickly and efficiently existing drugs can be evaluated for new purposes.<\/p>\n

“Think of it this way: Other methods try to design a key for a lock that’s sitting perfectly still, but in your body, that lock is constantly jiggling and changing shape. Our AI designs the key while the lock is moving, so the fit is much more realistic,” said Dokholyan, of UVA’s Department of Neurology. “This could make a real difference for patients with cancer, neurological disorders and many other conditions where we desperately need better drugs targeting these wiggly proteins but keep hitting dead ends.”<\/p>\n

Dokholyan and his team have described the development and results of these tools in papers in the journals Proceedings of the National Academy of Sciences<\/a>, Journal of Chemical Information and Modeling<\/a> and Science Advances<\/a>. The research team includes Wang, Dong Yan Zhang, Shreshty Budakoti and Dokholyan.<\/p>\n

\"New<\/p>\n

Workflow and architecture of YuelDesign. Credit: Proceedings of the National Academy of Sciences (2026). DOI: 10.1073\/pnas.2524913123<\/p>\n

\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\tThe pitfalls of drug development<\/p>\n

The average cost of developing a new drug has been estimated to reach or exceed $2.6 billion, and almost 90% of new drugs fail when they reach human testing. That is due, in no small part, to the difficulty of predicting how molecules in a drug will interact (bind) with their targets in the body. If a molecule doesn’t bind exactly as intended at exactly the right spot, the drug won’t work, or could have unwanted, harmful side effects.<\/p>\n

Artificial intelligence has helped address this problem, greatly accelerating drug design, but Dokholyan’s work takes it to the next level. His YuelDesign overcomes limitations of the existing options by designing drug molecules while treating proteins as flexible, dynamic structures, not the rigid and frozen snapshots used by other methods. This is critical because proteins often change shape<\/a> when a drug binds to them, a phenomenon known as “induced fit.” Ignoring this flexibility can lead to drugs that look promising on a computer screen but fail in reality.<\/p>\n

Dokholyan and his team designed YuelDesign specifically to overcome this problem. Using advanced AI “diffusion models<\/a>,” the technology simultaneously generates both the protein pocket structure and the small molecule that can slot into it\u2014the key that will turn the lock, allowing both to adapt to each other during the design process.<\/p>\n

\n Discover the latest in science, tech, and space with over 100,000 subscribers who rely on Phys.org for daily insights.
\n Sign up for our free newsletter<\/a> and get updates on breakthroughs,
\n innovations, and research that matter\u2014daily or weekly.\n <\/p>\n

A companion tool, YuelPocket, uses graph neural networks<\/a> to identify precisely where on a protein a drug should bind, even on predicted protein structures from existing tools such as AlphaFold.<\/p>\n

“Most existing AI tools treat the protein as a frozen statue, but that’s not how biology works. Our approach lets the protein and the drug candidate evolve together during the design process, just as they would in the body,” said researcher Dr. Jian Wang. “We showed, for example, that when designing molecules for a well-known cancer-related protein<\/a> called CDK2, only YuelDesign could capture the critical structural changes that happen when a drug binds.”<\/p>\n

Mapping out protein pockets is critical to “virtually every aspect of modern development,” the researchers note in a new scientific paper outlining their YuelPocket testing. The promising results have Dokholyan hopeful that the technology can reduce drug development costs, improve the success rate of new drug candidates and accelerate how quickly new treatments and cures can reach patients.<\/p>\n

“Our ultimate goal is to make drug discovery faster, cheaper and more likely to succeed, so that promising treatments can reach patients sooner,” Dokholyan said, adding that he wants to “democratize” drug discovery by putting new tools at scientists’ fingertips.<\/p>\n

“We’ve made all of our tools freely available to the scientific community. We want researchers anywhere in the world to be able to use them to tackle the diseases that matter most to their patients.”<\/p>\n

\t\t\t\t\t\t\t\t\t\t\t\t\t\tPublication details\t\t\t\t\t\t\t\t\t\t\t\t\t<\/p>\n

Jian Wang et al, Unified protein\u2013small molecule graph neural networks for binding site prediction, Proceedings of the National Academy of Sciences (2026). DOI: 10.1073\/pnas.2524913123<\/a><\/p>\n

Jian Wang et al, Multimodal Bond Reconstruction toward Generative Molecular Design, Journal of Chemical Information and Modeling (2026). DOI: 10.1021\/acs.jcim.5c03052<\/a><\/p>\n

Jian Wang et al, A Diffusion-Based Framework for Designing Molecules in Flexible Protein Pockets, Science Advances (2026). DOI: 10.1126\/sciadv.aeb7045<\/a>. www.science.org\/doi\/10.1126\/sciadv.aeb7045<\/a><\/p>\n

\t\t\t\t\t\t\t\t\t\t\t\tKey concepts
\n\t\t\t\t\t\t\t\t\t\t\t\tArtificial intelligence<\/a>\t\t\t\t\t\t\t\t\t\t\t<\/p>\n

\n\t\t\t\t\t\t\t\t\t\t\t\t\tProvided by
\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\tUniversity of Virginia<\/a>
\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/p>\n

\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/a>\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/p>\n

\n\t\t\t\t\t\t\t\t\t\t\t\tCitation:
\n\t\t\t\t\t\t\t\t\t\t\t\tAI diffusion models tailor drug molecules to custom-fit protein targets, speeding drug development and evaluation (2026, April 9)
\n\t\t\t\t\t\t\t\t\t\t\t\tretrieved 9 April 2026
\n\t\t\t\t\t\t\t\t\t\t\t\tfrom https:\/\/phys.org\/news\/2026-04-ai-diffusion-tailor-drug-molecules.html\n\t\t\t\t\t\t\t\t\t\t\t <\/p>\n

\n\t\t\t\t\t\t\t\t\t\t\t This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no
\n\t\t\t\t\t\t\t\t\t\t\t part may be reproduced without the written permission. The content is provided for information purposes only.\n\t\t\t\t\t\t\t\t\t\t\t <\/p>\n","protected":false},"excerpt":{"rendered":"Structural and topological representation of protein\u2013compound interactions in YuelPocket’s graph neural network framework. Credit: Proceedings of the National…\n","protected":false},"author":2,"featured_media":1998,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[2],"tags":[24,25,165,166,164,161,160,162,134,163],"class_list":{"0":"post-1997","1":"post","2":"type-post","3":"status-publish","4":"format-standard","5":"has-post-thumbnail","7":"category-ai","8":"tag-ai","9":"tag-artificial-intelligence","10":"tag-materials","11":"tag-nanotech","12":"tag-physics","13":"tag-physics-news","14":"tag-science","15":"tag-science-news","16":"tag-technology","17":"tag-technology-news"},"_links":{"self":[{"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/posts\/1997","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/comments?post=1997"}],"version-history":[{"count":0,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/posts\/1997\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/media\/1998"}],"wp:attachment":[{"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/media?parent=1997"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/categories?post=1997"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/tags?post=1997"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}