{"id":1188268,"date":"2026-10-01T14:33:43","date_gmt":"2026-10-01T21:33:43","guid":{"rendered":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/publication\/inverse-folddir-structure-conditioned-protein-sequence-design-by-dirichlet-flow-matching\/"},"modified":"2026-10-07T11:28:40","modified_gmt":"2026-10-07T18:28:40","slug":"inverse-folddir-structure-conditioned-protein-sequence-design-by-dirichlet-flow-matching","status":"publish","type":"msr-research-item","link":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/publication\/inverse-folddir-structure-conditioned-protein-sequence-design-by-dirichlet-flow-matching\/","title":{"rendered":"Inverse FoldDir: Structure-conditioned Protein Sequence Design by Dirichlet Flow Matching"},"content":{"rendered":"\n\n\n<p class=\"wp-block-paragraph\"><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Protein engineering has important implications in the bioeconomy, enabling applications in materials, medicine, and energy. A key challenge is designing protein sequences that have a specific form and function. Protein inverse folding seeks to address this challenge by identifying amino acid sequences compatible with a desired protein backbone. This task is central to protein redesign and can provide a sequence-design capability for de novo backbones produced by structure-generation methods. Ideally, inverse folding can provide diverse sequence alternatives, fixed residues or motifs, soft biochemical preferences at selected positions, and candidates that remain experimentally useful. We developed Inverse FoldDir, a controllable inverse-folding method that performs iterative denoising on the amino acid probability simplex. Given a backbone structure, the model updates all positions jointly through a learned Dirichlet flow, supporting full sequence generation, fixed-residue inpainting, and user-defined soft residue priors. On the held-out CATH 4.2 test set, Inverse FoldDir achieved a mean TM-score of 84.5 (on a 0-100 scale) and a mean C\u03b1 RMSD of 1.76 \u00c5, compared with 83.3 and 1.86 \u00c5, respectively, for ESM-IF1, the strongest evaluated baseline on both metrics. Denoising trajectory analyses showed that positions commit at different rates and that some residues change identity late in generation, illustrating whole-sequence refinement rather than one-shot prediction or irreversible sequential decoding. We experimentally tested Inverse FoldDir in an anti-GFP nanobody redesign task, where two of 35 redesigned sequences retained reproducible sfGFP-binding signal across independent assay runs with approximately 43% sequence divergence from the native nanobody. Inverse FoldDir is a structure-conditioned protein redesign method that combines structural recovery, user control, experimental validation, and a natural route toward future property-guided sampling.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>Protein engineering has important implications in the bioeconomy, enabling applications in materials, medicine, and energy. A key challenge is designing protein sequences that have a specific form and function. Protein inverse folding seeks to address this challenge by identifying amino acid sequences compatible with a desired protein backbone. This task is central to protein redesign [&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":[{"type":"text","value":"Alp Tartici","user_id":0},{"type":"text","value":"M. Stojkovic","user_id":0},{"type":"text","value":"An-Ru Tian","user_id":0},{"type":"text","value":"Michael C. Jewett","user_id":0},{"type":"text","value":"R. Altman","user_id":0},{"type":"user_nicename","value":"Bruce Wittmann","user_id":"43991"}],"msr_publishername":"","msr_publisher_other":"","msr_booktitle":"","msr_chapter":"","msr_edition":"","msr_editors":"","msr_how_published":"National Library of Medicine","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":"10.64898\/2026.09.06.749733","msr_arxiv_id":"","msr_mag_id":"","msr_other_authors":"","msr_other_contributors":"","msr_speaker":"","msr_award":"","msr_affiliation":"","msr_institution":"","msr_host":"","msr_version":"","msr_duration":"","msr_release_tracker_id":"","msr_highlight_type":"","msr_date_display_format":"","msr_main_download_label":"","msr_external_link_label":"","msr_doi_label":"","msr_published_date":"2026-09-11","msr_startdate":"","msr_presentation_date":"","msr_highlight_text":"","msr_notes":"","msr_longbiography":"","msr_publicationurl":"","msr_external_url":"","msr_secondary_video_url":"","msr_conference_url":"","msr_journal_url":"","msr_year":2026,"msr_month":9,"msr_day":11,"msr_microsoftintellectualproperty":false,"msr_pub_id":"84f494214788e8c8d13e89fbbc93f2ad35e26a3c","msr_publication_uploader":[{"type":"url","title":"https:\/\/www.ncbi.nlm.nih.gov\/pmc\/articles\/13622493","label_id":243109,"id":false,"viewUrl":false},{"type":"url","title":"https:\/\/www.biorxiv.org\/content\/biorxiv\/early\/2026\/09\/07\/2026.09.06.749733.full.pdf","label_id":243109,"id":false,"viewUrl":false}],"msr_related_uploader":[],"msr_original_fields_of_study":[],"msr_s2_paper_id":"84f494214788e8c8d13e89fbbc93f2ad35e26a3c","msr_s2_pdf_url":"https:\/\/www.biorxiv.org\/content\/biorxiv\/early\/2026\/09\/07\/2026.09.06.749733.full.pdf","msr_citation_count_updated":"","msr_citation_count":0,"msr_influential_citations":0,"msr_reference_count":32,"msr_s2_open_access":true,"msr_s2_author_ids":[],"msr_pub_ids":[{"provider":"s2","id":"84f494214788e8c8d13e89fbbc93f2ad35e26a3c"},{"provider":"doi","id":"10.64898\/2026.09.06.749733"},{"provider":"pubmedcentral","id":"13622493"},{"provider":"corpusid","id":"291876937"}],"msr_hide_image_in_river":0,"footnotes":""},"msr-research-highlight":[],"research-area":[13556,13553],"msr-publication-type":[270373],"msr-publisher":[],"msr-publication-cta":[],"msr-focus-area":[],"msr-locale":[268875],"msr-post-option":[],"msr-field-of-study":[250972,251227,251278],"msr-conference":[],"msr-journal":[259099],"msr-impact-theme":[],"msr-pillar":[],"class_list":["post-1188268","msr-research-item","type-msr-research-item","status-publish","hentry","msr-research-area-artificial-intelligence","msr-research-area-medical-health-genomics","msr-locale-en_us","msr-field-of-study-biology","msr-field-of-study-computational-biology","msr-field-of-study-protein-engineering"],"msr_publishername":"","msr_edition":"","msr_affiliation":"","msr_published_date":"2026-09-11","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":"National Library of Medicine","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":0,"msr_main_download":"","msr_publicationurl":"","msr_doi":"10.64898\/2026.09.06.749733","msr_publication_uploader":[{"type":"url","title":"https:\/\/www.ncbi.nlm.nih.gov\/pmc\/articles\/13622493","label_id":243109,"id":false,"viewUrl":false},{"type":"url","title":"https:\/\/www.biorxiv.org\/content\/biorxiv\/early\/2026\/09\/07\/2026.09.06.749733.full.pdf","label_id":243109,"id":false,"viewUrl":false}],"msr_related_uploader":[],"msr_citation_count":0,"msr_citation_count_updated":"","msr_s2_paper_id":"84f494214788e8c8d13e89fbbc93f2ad35e26a3c","msr_influential_citations":0,"msr_reference_count":32,"msr_arxiv_id":"","msr_s2_author_ids":[],"msr_s2_open_access":true,"msr_s2_pdf_url":"https:\/\/www.biorxiv.org\/content\/biorxiv\/early\/2026\/09\/07\/2026.09.06.749733.full.pdf","msr_attachments":[],"msr-author-ordering":[{"type":"text","value":"Alp Tartici","user_id":0,"rest_url":false},{"type":"text","value":"M. Stojkovic","user_id":0,"rest_url":false},{"type":"text","value":"An-Ru Tian","user_id":0,"rest_url":false},{"type":"text","value":"Michael C. Jewett","user_id":0,"rest_url":false},{"type":"text","value":"R. Altman","user_id":0,"rest_url":false},{"type":"user_nicename","value":"Bruce Wittmann","user_id":43991,"rest_url":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Bruce Wittmann"}],"msr_impact_theme":[],"msr_research_lab":[],"msr_event":[],"msr_group":[],"msr_project":[],"publication":[],"video":[],"msr-tool":[],"msr_publication_type":"misc","related_content":[],"_links":{"self":[{"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/1188268","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item"}],"about":[{"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/types\/msr-research-item"}],"version-history":[{"count":3,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/1188268\/revisions"}],"predecessor-version":[{"id":1188947,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/1188268\/revisions\/1188947"}],"wp:attachment":[{"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/media?parent=1188268"}],"wp:term":[{"taxonomy":"msr-research-highlight","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-research-highlight?post=1188268"},{"taxonomy":"msr-research-area","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/research-area?post=1188268"},{"taxonomy":"msr-publication-type","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-publication-type?post=1188268"},{"taxonomy":"msr-publisher","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-publisher?post=1188268"},{"taxonomy":"msr-publication-cta","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-publication-cta?post=1188268"},{"taxonomy":"msr-focus-area","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-focus-area?post=1188268"},{"taxonomy":"msr-locale","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-locale?post=1188268"},{"taxonomy":"msr-post-option","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-post-option?post=1188268"},{"taxonomy":"msr-field-of-study","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-field-of-study?post=1188268"},{"taxonomy":"msr-conference","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-conference?post=1188268"},{"taxonomy":"msr-journal","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-journal?post=1188268"},{"taxonomy":"msr-impact-theme","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-impact-theme?post=1188268"},{"taxonomy":"msr-pillar","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-pillar?post=1188268"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}