Following the sequence and structure revolutions, predicting functionally relevant protein structure changes at scale remains an outstanding challenge. We introduce BioEmu, a deep learning system that emulates protein equilibrium ensembles by generating thousands of statistically independent structures per hour on a single GPU. BioEmu integrates over 200 milliseconds of molecular dynamics (MD) simulations, static structures and experimental protein stabilities using novel training algorithms. It captures diverse functional motions\u2014including cryptic pocket formation, local unfolding, and domain rearrangements\u2014and predicts relative free energies with 1 kcal\/mol accuracy compared to millisecond-scale MD and experimental data. BioEmu provides mechanistic insights by jointly modelling structural ensembles and thermodynamic properties. This approach amortizes the cost of MD and experimental data generation, demonstrating a scalable path toward understanding and designing protein function.<\/p>\n","protected":false},"excerpt":{"rendered":"
Following the sequence and structure revolutions, predicting functionally relevant protein structure changes at scale remains an outstanding challenge. We introduce BioEmu, a deep learning system that emulates protein equilibrium ensembles by generating thousands of statistically independent structures per hour on a single GPU. BioEmu integrates over 200 milliseconds of molecular dynamics (MD) simulations, static structures […]<\/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":"Science","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":"2025-7-10","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":[193715],"msr-publisher":[],"msr-focus-area":[],"msr-locale":[268875],"msr-post-option":[269148,269142,269145],"msr-field-of-study":[250972],"msr-conference":[],"msr-journal":[],"msr-impact-theme":[],"msr-pillar":[],"class_list":["post-1115088","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-post-option-pinned-for-river","msr-field-of-study-biology"],"msr_publishername":"","msr_edition":"","msr_affiliation":"","msr_published_date":"2025-7-10","msr_host":"","msr_duration":"","msr_version":"","msr_speaker":"","msr_other_contributors":"","msr_booktitle":"","msr_pages_string":"","msr_chapter":"","msr_isbn":"","msr_journal":"Science","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:\/\/www.science.org\/doi\/10.1126\/science.adv9817","label_id":"243109","label":0},{"type":"file","viewUrl":"https:\/\/newed.any0.dpdns.org\/en-us\/research\/wp-content\/uploads\/2024\/12\/accepted_paper.pdf","id":"1144496","title":"accepted_paper","label_id":"243132","label":0}],"msr_related_uploader":[{"type":"url","viewUrl":"false","id":"false","title":"https:\/\/github.com\/microsoft\/bioemu","label_id":"264520","label":0}],"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":[{"id":1144496,"url":"https:\/\/newed.any0.dpdns.org\/en-us\/research\/wp-content\/uploads\/2024\/12\/accepted_paper.pdf"}],"msr-author-ordering":[{"type":"user_nicename","value":"Sarah Lewis","user_id":41305,"rest_url":"https:\/\/newed.any0.dpdns.org\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Sarah Lewis"},{"type":"user_nicename","value":"Tim Hempel","user_id":43470,"rest_url":"https:\/\/newed.any0.dpdns.org\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Tim Hempel"},{"type":"user_nicename","value":"Jose Jimenez-Luna","user_id":41515,"rest_url":"https:\/\/newed.any0.dpdns.org\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Jose Jimenez-Luna"},{"type":"text","value":"Michael Gastegger","user_id":0,"rest_url":false},{"type":"user_nicename","value":"Yu Xie","user_id":43047,"rest_url":"https:\/\/newed.any0.dpdns.org\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Yu Xie"},{"type":"user_nicename","value":"Andrew Foong","user_id":42807,"rest_url":"https:\/\/newed.any0.dpdns.org\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Andrew Foong"},{"type":"user_nicename","value":"Victor Garc\u00eda Satorras","user_id":41832,"rest_url":"https:\/\/newed.any0.dpdns.org\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Victor Garc\u00eda Satorras"},{"type":"text","value":"Osama Abdin","user_id":0,"rest_url":false},{"type":"user_nicename","value":"Bas Veeling","user_id":41916,"rest_url":"https:\/\/newed.any0.dpdns.org\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Bas Veeling"},{"type":"text","value":"I. Zaporozhets","user_id":0,"rest_url":false},{"type":"text","value":"Yaoyi Chen","user_id":0,"rest_url":false},{"type":"text","value":"Soojung Yang","user_id":0,"rest_url":false},{"type":"text","value":"Arne Schneuing","user_id":0,"rest_url":false},{"type":"text","value":"Jigyasa Nigam","user_id":0,"rest_url":false},{"type":"text","value":"Federico Barbero","user_id":0,"rest_url":false},{"type":"text","value":"Vincent Stimper","user_id":0,"rest_url":false},{"type":"text","value":"Andrew Campbell","user_id":0,"rest_url":false},{"type":"text","value":"Jason Yim","user_id":0,"rest_url":false},{"type":"text","value":"Marten Lienen","user_id":0,"rest_url":false},{"type":"user_nicename","value":"Yu Shi","user_id":40405,"rest_url":"https:\/\/newed.any0.dpdns.org\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Yu Shi"},{"type":"user_nicename","value":"Shuxin Zheng","user_id":38523,"rest_url":"https:\/\/newed.any0.dpdns.org\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Shuxin Zheng"},{"type":"user_nicename","value":"Hannes Schulz","user_id":37188,"rest_url":"https:\/\/newed.any0.dpdns.org\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Hannes Schulz"},{"type":"user_nicename","value":"Usman Munir","user_id":42582,"rest_url":"https:\/\/newed.any0.dpdns.org\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Usman Munir"},{"type":"user_nicename","value":"Cecilia Clementi","user_id":42501,"rest_url":"https:\/\/newed.any0.dpdns.org\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Cecilia Clementi"},{"type":"user_nicename","value":"Frank No\u00e9","user_id":42216,"rest_url":"https:\/\/newed.any0.dpdns.org\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Frank No\u00e9"}],"msr_impact_theme":[],"msr_research_lab":[851467],"msr_event":[],"msr_group":[],"msr_project":[1149439],"publication":[],"video":[],"msr-tool":[1143073],"msr_publication_type":"article","related_content":{"projects":[{"ID":1149439,"post_title":"Biomolecules","post_name":"biomolecules","post_type":"msr-project","post_date":"2025-10-08 08:00:00","post_modified":"2026-01-26 08:00:29","post_status":"publish","permalink":"https:\/\/newed.any0.dpdns.org\/en-us\/research\/project\/biomolecules\/","post_excerpt":"Following the sequence and structure revolutions, predicting functionally relevant protein structure changes at scale remains an outstanding challenge. BioEmu is a deep learning system that emulates protein equilibrium ensembles by generating thousands of statistically independent structures per hour on a single GPU. BioEmu integrates over 200 milliseconds of molecular dynamics (MD) simulations, static structures and experimental protein stabilities using novel training algorithms. It captures diverse functional motions \u2013 including cryptic pocket formation, local unfolding, and…","_links":{"self":[{"href":"https:\/\/newed.any0.dpdns.org\/en-us\/research\/wp-json\/wp\/v2\/msr-project\/1149439"}]}}]},"_links":{"self":[{"href":"https:\/\/newed.any0.dpdns.org\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/1115088","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/newed.any0.dpdns.org\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item"}],"about":[{"href":"https:\/\/newed.any0.dpdns.org\/en-us\/research\/wp-json\/wp\/v2\/types\/msr-research-item"}],"version-history":[{"count":3,"href":"https:\/\/newed.any0.dpdns.org\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/1115088\/revisions"}],"predecessor-version":[{"id":1144497,"href":"https:\/\/newed.any0.dpdns.org\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/1115088\/revisions\/1144497"}],"wp:attachment":[{"href":"https:\/\/newed.any0.dpdns.org\/en-us\/research\/wp-json\/wp\/v2\/media?parent=1115088"}],"wp:term":[{"taxonomy":"msr-research-highlight","embeddable":true,"href":"https:\/\/newed.any0.dpdns.org\/en-us\/research\/wp-json\/wp\/v2\/msr-research-highlight?post=1115088"},{"taxonomy":"msr-research-area","embeddable":true,"href":"https:\/\/newed.any0.dpdns.org\/en-us\/research\/wp-json\/wp\/v2\/research-area?post=1115088"},{"taxonomy":"msr-publication-type","embeddable":true,"href":"https:\/\/newed.any0.dpdns.org\/en-us\/research\/wp-json\/wp\/v2\/msr-publication-type?post=1115088"},{"taxonomy":"msr-publisher","embeddable":true,"href":"https:\/\/newed.any0.dpdns.org\/en-us\/research\/wp-json\/wp\/v2\/msr-publisher?post=1115088"},{"taxonomy":"msr-focus-area","embeddable":true,"href":"https:\/\/newed.any0.dpdns.org\/en-us\/research\/wp-json\/wp\/v2\/msr-focus-area?post=1115088"},{"taxonomy":"msr-locale","embeddable":true,"href":"https:\/\/newed.any0.dpdns.org\/en-us\/research\/wp-json\/wp\/v2\/msr-locale?post=1115088"},{"taxonomy":"msr-post-option","embeddable":true,"href":"https:\/\/newed.any0.dpdns.org\/en-us\/research\/wp-json\/wp\/v2\/msr-post-option?post=1115088"},{"taxonomy":"msr-field-of-study","embeddable":true,"href":"https:\/\/newed.any0.dpdns.org\/en-us\/research\/wp-json\/wp\/v2\/msr-field-of-study?post=1115088"},{"taxonomy":"msr-conference","embeddable":true,"href":"https:\/\/newed.any0.dpdns.org\/en-us\/research\/wp-json\/wp\/v2\/msr-conference?post=1115088"},{"taxonomy":"msr-journal","embeddable":true,"href":"https:\/\/newed.any0.dpdns.org\/en-us\/research\/wp-json\/wp\/v2\/msr-journal?post=1115088"},{"taxonomy":"msr-impact-theme","embeddable":true,"href":"https:\/\/newed.any0.dpdns.org\/en-us\/research\/wp-json\/wp\/v2\/msr-impact-theme?post=1115088"},{"taxonomy":"msr-pillar","embeddable":true,"href":"https:\/\/newed.any0.dpdns.org\/en-us\/research\/wp-json\/wp\/v2\/msr-pillar?post=1115088"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}