{"id":1157655,"date":"2025-12-05T05:54:31","date_gmt":"2025-12-05T13:54:31","guid":{"rendered":"https:\/\/www.microsoft.com\/en-us\/research\/?post_type=msr-research-item&#038;p=1157655"},"modified":"2025-12-05T05:56:26","modified_gmt":"2025-12-05T13:56:26","slug":"accelerating-biomolecular-modeling-with-atomworks-and-rf3","status":"publish","type":"msr-research-item","link":"https:\/\/www.microsoft.com\/en-us\/research\/publication\/accelerating-biomolecular-modeling-with-atomworks-and-rf3\/","title":{"rendered":"Accelerating Biomolecular Modeling with AtomWorks and RF3"},"content":{"rendered":"<p>Deep learning methods trained on protein structure databases have revolutionized biomolecular structure prediction, but developing and training new models remains a considerable challenge. To facilitate the development of new models, we present AtomWorks: a broadly applicable data framework for developing state-of-the-art biomolecular foundation models spanning diverse tasks, including structure prediction, generative protein design, and fixed backbone sequence design. We use AtomWorks to train RosettaFold-3 (RF3), a structure prediction network capable of predicting arbitrary biomolecular complexes with an improved treatment of chirality that narrows the performance gap between closed-source AlphaFold3 (AF3) and existing open-source implementations. We expect that AtomWorks will accelerate the next generation of open-source biomolecular machine learning models and that RF3 will be broadly useful as a structure prediction tool. To this end, we release the AtomWorks framework (<a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" href=\"https:\/\/github.com\/RosettaCommons\/atomworks\" target=\"_blank\" rel=\"noopener noreferrer\">https:\/\/github.com\/RosettaCommons\/atomworks<span class=\"sr-only\"> (opens in new tab)<\/span><\/a>), together with curated training data, code and model weights for RF3 (<a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" href=\"https:\/\/github.com\/RosettaCommons\/modelforge\" target=\"_blank\" rel=\"noopener noreferrer\">https:\/\/github.com\/RosettaCommons\/modelforge<span class=\"sr-only\"> (opens in new tab)<\/span><\/a>) under a permissive BSD license.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>Deep learning methods trained on protein structure databases have revolutionized biomolecular structure prediction, but developing and training new models remains a considerable challenge. To facilitate the development of new models, we present AtomWorks: a broadly applicable data framework for developing state-of-the-art biomolecular foundation models spanning diverse tasks, including structure prediction, generative protein design, and fixed [&hellip;]<\/p>\n","protected":false},"featured_media":0,"template":"","meta":{"msr-url-field":"","msr-podcast-episode":"","msrModifiedDate":"","msrModifiedDateEnabled":false,"ep_exclude_from_search":false,"_classifai_error":"","msr-author-ordering":null,"msr_publishername":"","msr_publisher_other":"","msr_booktitle":"","msr_chapter":"","msr_edition":"","msr_editors":"","msr_how_published":"","msr_isbn":"","msr_issue":"","msr_journal":"","msr_number":"","msr_organization":"","msr_pages_string":"","msr_page_range_start":"","msr_page_range_end":"","msr_series":"","msr_volume":"","msr_copyright":"","msr_conference_name":"","msr_doi":"","msr_arxiv_id":"","msr_s2_paper_id":"","msr_mag_id":"","msr_pubmed_id":"","msr_other_authors":"","msr_other_contributors":"","msr_speaker":"","msr_award":"","msr_affiliation":"","msr_institution":"","msr_host":"","msr_version":"","msr_duration":"","msr_original_fields_of_study":null,"msr_release_tracker_id":"","msr_s2_match_type":"","msr_citation_count_updated":"","msr_published_date":"2025-8-14","msr_highlight_text":"","msr_notes":"","msr_longbiography":"","msr_publicationurl":"","msr_external_url":"","msr_secondary_video_url":"","msr_conference_url":"","msr_journal_url":"","msr_s2_pdf_url":"","msr_year":0,"msr_citation_count":0,"msr_influential_citations":0,"msr_reference_count":0,"msr_s2_match_confidence":0,"msr_microsoftintellectualproperty":true,"msr_s2_open_access":false,"msr_s2_author_ids":[],"msr_pub_ids":[],"msr_hide_image_in_river":null,"footnotes":""},"msr-research-highlight":[],"research-area":[13556],"msr-publication-type":[193726],"msr-publisher":[],"msr-focus-area":[],"msr-locale":[268875],"msr-post-option":[269148,269142],"msr-field-of-study":[250972,246985],"msr-conference":[],"msr-journal":[],"msr-impact-theme":[],"msr-pillar":[],"class_list":["post-1157655","msr-research-item","type-msr-research-item","status-publish","hentry","msr-research-area-artificial-intelligence","msr-locale-en_us","msr-post-option-approved-for-river","msr-post-option-include-in-river","msr-field-of-study-biology","msr-field-of-study-medicine"],"msr_publishername":"","msr_edition":"","msr_affiliation":"","msr_published_date":"2025-8-14","msr_host":"","msr_duration":"","msr_version":"","msr_speaker":"","msr_other_contributors":"","msr_booktitle":"","msr_pages_string":"","msr_chapter":"","msr_isbn":"","msr_journal":"","msr_volume":"","msr_number":"","msr_editors":"","msr_series":"","msr_issue":"","msr_organization":"","msr_how_published":"","msr_notes":"","msr_highlight_text":"","msr_release_tracker_id":"","msr_original_fields_of_study":"","msr_download_urls":"","msr_external_url":"","msr_secondary_video_url":"","msr_longbiography":"","msr_microsoftintellectualproperty":1,"msr_main_download":"","msr_publicationurl":"","msr_doi":"","msr_publication_uploader":[{"type":"doi","viewUrl":"false","id":"false","title":"https:\/\/doi.org\/10.1101\/2025.08.14.670328","label_id":"252679","label":0},{"type":"url","viewUrl":"false","id":"false","title":"https:\/\/pubmed.ncbi.nlm.nih.gov\/40832246","label_id":"252679","label":0}],"msr_related_uploader":"","msr_citation_count":0,"msr_citation_count_updated":"","msr_s2_paper_id":"","msr_influential_citations":0,"msr_reference_count":0,"msr_arxiv_id":"","msr_s2_author_ids":[],"msr_s2_open_access":false,"msr_s2_pdf_url":null,"msr_attachments":[],"msr-author-ordering":[{"type":"text","value":"Nathaniel Corley","user_id":0,"rest_url":false},{"type":"text","value":"Simon V. Mathis","user_id":0,"rest_url":false},{"type":"text","value":"Rohith Krishna","user_id":0,"rest_url":false},{"type":"text","value":"Magnus Bauer","user_id":0,"rest_url":false},{"type":"text","value":"Tuscan R. Thompson","user_id":0,"rest_url":false},{"type":"text","value":"Woody Ahern","user_id":0,"rest_url":false},{"type":"text","value":"Maxwell W. Kazman","user_id":0,"rest_url":false},{"type":"text","value":"Rafael I Brent","user_id":0,"rest_url":false},{"type":"text","value":"Kieran Didi","user_id":0,"rest_url":false},{"type":"text","value":"Andrew Kubaney","user_id":0,"rest_url":false},{"type":"text","value":"Lilian McHugh","user_id":0,"rest_url":false},{"type":"text","value":"Arnav Nagle","user_id":0,"rest_url":false},{"type":"text","value":"Andrew Favor","user_id":0,"rest_url":false},{"type":"user_nicename","value":"Meghana Kshirsagar","user_id":39736,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Meghana Kshirsagar"},{"type":"text","value":"Pascal Sturmfels","user_id":0,"rest_url":false},{"type":"text","value":"Yanjing Li","user_id":0,"rest_url":false},{"type":"text","value":"J. Butcher","user_id":0,"rest_url":false},{"type":"text","value":"Bo Qiang","user_id":0,"rest_url":false},{"type":"text","value":"Lars L. Schaaf","user_id":0,"rest_url":false},{"type":"text","value":"Raktim Mitra","user_id":0,"rest_url":false},{"type":"text","value":"Katelyn V. Campbell","user_id":0,"rest_url":false},{"type":"text","value":"Odin Zhang","user_id":0,"rest_url":false},{"type":"text","value":"Roni Weissman","user_id":0,"rest_url":false},{"type":"text","value":"Ian R. Humphreys","user_id":0,"rest_url":false},{"type":"text","value":"Qian Cong","user_id":0,"rest_url":false},{"type":"text","value":"Jonathan Funk","user_id":0,"rest_url":false},{"type":"text","value":"Shreyash Sonthalia","user_id":0,"rest_url":false},{"type":"text","value":"Pietro Lio","user_id":0,"rest_url":false},{"type":"text","value":"David Baker","user_id":0,"rest_url":false},{"type":"text","value":"F. DiMaio","user_id":0,"rest_url":false}],"msr_impact_theme":[],"msr_research_lab":[],"msr_event":[],"msr_group":[696544],"msr_project":[],"publication":[],"video":[],"msr-tool":[1157650],"msr_publication_type":"unpublished","related_content":[],"_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/1157655","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item"}],"about":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/types\/msr-research-item"}],"version-history":[{"count":3,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/1157655\/revisions"}],"predecessor-version":[{"id":1157658,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/1157655\/revisions\/1157658"}],"wp:attachment":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media?parent=1157655"}],"wp:term":[{"taxonomy":"msr-research-highlight","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-highlight?post=1157655"},{"taxonomy":"msr-research-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/research-area?post=1157655"},{"taxonomy":"msr-publication-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-publication-type?post=1157655"},{"taxonomy":"msr-publisher","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-publisher?post=1157655"},{"taxonomy":"msr-focus-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-focus-area?post=1157655"},{"taxonomy":"msr-locale","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-locale?post=1157655"},{"taxonomy":"msr-post-option","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-post-option?post=1157655"},{"taxonomy":"msr-field-of-study","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-field-of-study?post=1157655"},{"taxonomy":"msr-conference","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-conference?post=1157655"},{"taxonomy":"msr-journal","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-journal?post=1157655"},{"taxonomy":"msr-impact-theme","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-impact-theme?post=1157655"},{"taxonomy":"msr-pillar","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-pillar?post=1157655"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}