{"version":1,"type":"story","url":"https://digestai.news/story/idpforge-presents-deep-learning-method-for-intrinsically-disordered-pr","json":"https://digestai.news/story/idpforge-presents-deep-learning-method-for-intrinsically-disordered-pr.json","markdown":"https://digestai.news/story/idpforge-presents-deep-learning-method-for-intrinsically-disordered-pr.md","slug":"idpforge-presents-deep-learning-method-for-intrinsically-disordered-pr","headline":"IDPForge presents deep learning method for intrinsically disordered proteins","summary":"IDPForge is a new machine‑learning approach that uses a transformer protein‑language diffusion model to generate all‑atom ensembles of intrinsically disordered proteins (IDPs) and disordered regions (IDRs). Unlike AlphaFold and RoseTTAFold, which struggle with IDPs, IDPForge does not need sequence‑specific training, back‑transformations from coarse‑grained models, or ensemble reweighting. The generated conformational ensembles agree well with NMR and SAXS solution data, and the tool allows optional biasing with experimental restraints.\n\nThe authors note that IDPForge is available as an open‑source resource, enabling integrative structural studies of proteins that contain disordered segments. Funding comes from the National Institutes of Health (grant 5R01GM12762706), NIH 1R35GM163971, the Canada Research Chairs Program, and the U.C. Berkeley Chemical Biology Training Program (NIH T32 GM152983). No competing interests were declared.\n\nBy extending AI protein modeling to disordered regions, IDPForge could improve drug target identification and accelerate biomedical research that relies on accurate protein ensembles.","keyPoints":["IDPForge uses a transformer diffusion model to generate IDP ensembles","No sequence‑specific training or back‑transformations required","Ensembles match NMR and SAXS experimental data"],"whyItMatters":"IDPForge extends AI protein modeling to disordered regions, improving drug target understanding and accelerating biomedical research.","category":{"slug":"research","name":"Research","url":"https://digestai.news/category/research"},"entities":{"companies":[],"models":["IDPForge"],"people":["S. De Castro","O. Zhang","Z.H. Liu","J.D.F.-K.","S.D.C.","T.H-G."]},"firstPublishedAt":"2026-10-30T00:00:00Z","updatedAt":"2026-10-30T00:00:00Z","sourceCount":1,"hasPrimarySource":true,"sources":[{"outlet":"Nature Machine Learning","title":"IDPForge: deep learning of proteins with global and local regions of disorder","url":"https://nature.com/articles/s41467-026-79191-4","publishedAt":"2026-10-30T00:00:00Z","type":"primary","primary":true,"lead":true}],"sourceNotes":null,"discussions":[],"thread":null,"cite":{"text":"Digest AI, \"IDPForge presents deep learning method for intrinsically disordered proteins\", 30 October 2026, https://digestai.news/story/idpforge-presents-deep-learning-method-for-intrinsically-disordered-pr","publisher":"Digest AI","title":"IDPForge presents deep learning method for intrinsically disordered proteins","datePublished":"2026-10-30T00:00:00Z","url":"https://digestai.news/story/idpforge-presents-deep-learning-method-for-intrinsically-disordered-pr"},"generatedBy":"Written by Digest AI's editorial model from the linked sources; the sources are the record.","license":"Headlines, digests and key points are written by Digest AI and may be quoted with a link to the story page. Linked articles belong to their publishers. Terms: https://digestai.news/terms#reuse"}