{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,10]],"date-time":"2026-08-10T05:30:44Z","timestamp":1786339844298,"version":"build-2736575974"},"reference-count":29,"publisher":"Wiley","issue":"7","license":[{"start":{"date-parts":[[2026,6,7]],"date-time":"2026-06-07T00:00:00Z","timestamp":1780790400000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/onlinelibrary.wiley.com\/termsAndConditions#vor"},{"start":{"date-parts":[[2026,6,7]],"date-time":"2026-06-07T00:00:00Z","timestamp":1780790400000},"content-version":"tdm","delay-in-days":0,"URL":"http:\/\/doi.wiley.com\/10.1002\/tdm_license_1.1"}],"funder":[{"DOI":"10.13039\/100004339","name":"Sanofi","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100004339","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Protein Science"],"published-print":{"date-parts":[[2026,7]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Predicting antibody and NANOBODY\u00ae VHH\u2013antigen complexes remains a critical challenge for state\u2010of\u2010the\u2010art structure prediction models, limiting their impact in therapeutic discovery pipelines. We introduce SNAC\u2010DB, an ML\u2010ready database and curation pipeline enriched with structural biology expertise, designed to accelerate model accuracy and generalization by providing 31\u201337% expanded structural diversity over existing resources like SAbDab through comprehensive re\u2010curation that extracts maximum value from available experimental structures. SNAC\u2010DB expands coverage by capturing often\u2010overlooked complexes and accurately identifying complete multi\u2010chain epitopes through improved biological\u2010assembly based logic. Built for ML practitioners, SNAC\u2010DB provides standardized formats with multi\u2010threshold structure\u2010based clustering to enable principled sample weighting during training. Using a rigorous benchmark of public PDB entries deposited post\u2010May 2024 plus confidential therapeutic structures, we evaluate seven leading models (Protenix\u2010v1, OpenFold\u20103p2, RosettaFold\u20103, Boltz\u20102, Boltz\u20101x, Chai\u20101, and AlphaFold2.3\u2010multimer) with evaluation methodology tailored to antibody\/NANOBODY\u00ae VHH\u2013antigen complexes to ensure correct handling of multi\u2010chain epitopes, revealing systematic performance gaps: success rates rarely exceed 25%, confidence\u2010based ranking fails to identify best predictions even when accurate structures exist in ensembles, and all models consistently struggle with therapeutically relevant NANOBODY\u00ae VHHs. Systematic evaluation of sampling strategies demonstrates that while generating 1000 samples per target substantially increases the likelihood of producing accurate structures (oracle selection improves from 11.9% to 50.5%), confidence\u2010based ranking remains nearly flat (between 10.9% and 14.9%), revealing that improved ranking mechanisms represent a more tractable path to performance gains. Finally, fine\u2010tuning GeoDock on SNAC\u2010DB yields higher success rates than training on SAbDab (11.0% vs. 7.1% for antibodies; 7.0% vs. 4.0% for NANOBODY\u00ae VHHs), suggesting that SNAC\u2010DB's expanded structural diversity translates to improved model generalization.<\/jats:p>","DOI":"10.1002\/pro.70655","type":"journal-article","created":{"date-parts":[[2026,6,8]],"date-time":"2026-06-08T05:46:17Z","timestamp":1780897577000},"update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["<scp>SNAC<\/scp>\n                    \u2010\n                    <scp>DB<\/scp>\n                    : An\n                    <scp>ML<\/scp>\n                    \u2010ready database for antibody and\n                    <scp>NANOBODY<\/scp>\n                    \u00ae\n                    <scp>VHH<\/scp>\n                    \u2013antigen complexes with expanded structural diversity and real\u2010world benchmarking"],"prefix":"10.1002","volume":"35","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-9197-0736","authenticated-orcid":false,"given":"Abhinav","family":"Gupta","sequence":"first","affiliation":[{"name":"Large Molecule Research Sanofi  Cambridge Massachusetts USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Bryan Munoz","family":"Rivero","sequence":"additional","affiliation":[{"name":"Large Molecule Research Sanofi  Cambridge Massachusetts USA"},{"name":"Department of Chemical and Biomolecular Engineering Johns Hopkins University  Baltimore Maryland USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ruijiang","family":"Li","sequence":"additional","affiliation":[{"name":"Large Molecule Research Sanofi  Cambridge Massachusetts USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jorge","family":"Roel\u2010Touris","sequence":"additional","affiliation":[{"name":"Large Molecule Research Sanofi  Barcelona Spain"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yves Fomekong","family":"Nanfack","sequence":"additional","affiliation":[{"name":"Large Molecule Research Sanofi  Cambridge Massachusetts USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Maria","family":"Wendt","sequence":"additional","affiliation":[{"name":"Large Molecule Research Sanofi  Cambridge Massachusetts USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yu","family":"Qiu","sequence":"additional","affiliation":[{"name":"Large Molecule Research Sanofi  Cambridge Massachusetts USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Norbert","family":"Furtmann","sequence":"additional","affiliation":[{"name":"Large Molecule Research Sanofi  Frankfurt Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2026,6,7]]},"reference":[{"key":"e_1_2_9_2_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-024-07487-w"},{"key":"e_1_2_9_3_1","doi-asserted-by":"publisher","DOI":"10.1093\/nar\/28.1.235"},{"key":"e_1_2_9_4_1","doi-asserted-by":"publisher","DOI":"10.1093\/nar\/gkae1091"},{"key":"e_1_2_9_5_1","article-title":"Chai\u20101: decoding the molecular interactions of life","author":"Chai Discovery","journal-title":"bioRxiv"},{"key":"e_1_2_9_6_1","doi-asserted-by":"publisher","DOI":"10.1002\/pro.4862"},{"key":"e_1_2_9_7_1","article-title":"Accelerating biomolecular modeling with atomworks and RF3","author":"Corley N","year":"2025","journal-title":"bioRxiv"},{"key":"e_1_2_9_8_1","doi-asserted-by":"publisher","DOI":"10.1093\/bioinformatics\/btv552"},{"key":"e_1_2_9_9_1","doi-asserted-by":"publisher","DOI":"10.1093\/nar\/gkt1043"},{"key":"e_1_2_9_10_1","article-title":"Protein complex prediction with AlphaFold\u2010Multimer","author":"Evans R","year":"2021","journal-title":"biorxiv"},{"key":"e_1_2_9_11_1","doi-asserted-by":"publisher","DOI":"10.1093\/bioinformatics\/btz517"},{"key":"e_1_2_9_12_1","doi-asserted-by":"publisher","DOI":"10.7554\/eLife.92462.3"},{"key":"e_1_2_9_13_1","unstructured":"HarprechtC.Structural analysis of JC polyomavirus inhibitors and receptors. PhD thesis Universit\u00e4t T\u00fcbingen.2020."},{"key":"e_1_2_9_14_1","doi-asserted-by":"publisher","DOI":"10.1002\/pro.70208"},{"key":"e_1_2_9_15_1","doi-asserted-by":"publisher","DOI":"10.1080\/19420862.2025.2545601"},{"key":"e_1_2_9_16_1","doi-asserted-by":"publisher","DOI":"10.1038\/nature19946"},{"key":"e_1_2_9_17_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-021-03819-2"},{"key":"e_1_2_9_18_1","unstructured":"JumperJ EvansR PritzelA GreenT FigurnovM et al.AlphaFold 2: code release.2021https:\/\/github.com\/deepmind\/alphafold"},{"key":"e_1_2_9_19_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41592-025-02593-7"},{"key":"e_1_2_9_20_1","doi-asserted-by":"publisher","DOI":"10.1093\/procel\/pwae073"},{"key":"e_1_2_9_21_1","doi-asserted-by":"publisher","DOI":"10.1093\/nar\/gkab1136"},{"key":"e_1_2_9_22_1","doi-asserted-by":"publisher","DOI":"10.1093\/bioinformatics\/btae586"},{"key":"e_1_2_9_23_1","article-title":"Boltz\u20102: towards accurate and efficient binding affinity prediction","author":"Passaro S","year":"2025","journal-title":"BioRxiv"},{"key":"e_1_2_9_24_1","article-title":"Protenix\u2010v1: toward high\u2010accuracy open\u2010source biomolecular structure prediction","author":"Protenix Team","year":"2026","journal-title":"bioRxiv"},{"key":"e_1_2_9_25_1","unstructured":"RCSB Protein Data Bank.Guide to understanding pdb data: Biological assemblies.2025https:\/\/pdb101.rcsb.org\/learn\/guide-to-understanding-pdb-data\/biological-assemblies"},{"key":"e_1_2_9_26_1","unstructured":"The OpenFold3 Team.OpenFold3\u2010preview.2025a."},{"key":"e_1_2_9_27_1","unstructured":"The OpenFold3 Team.OpenFold3\u2010preview2 technical report. Technical report Columbia University and Lawrence Livermore National Laboratory and Novo Nordisk and AWS and Seoul National University and Chan Zuckerberg Initiative and Absci and SandboxAQ and AMD and NVIDIA and University of Bristol and OpenFold Consortium. Preprint.2025b."},{"key":"e_1_2_9_28_1","doi-asserted-by":"publisher","DOI":"10.1093\/bib\/bbag137"},{"key":"e_1_2_9_29_1","article-title":"Boltz\u20101: democratizing biomolecular interaction modeling","author":"Wohlwend J","year":"2024","journal-title":"bioRxiv"},{"key":"e_1_2_9_30_1","doi-asserted-by":"publisher","DOI":"10.1002\/pro.4865"}],"container-title":["Protein Science"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/pro.70655","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/full-xml\/10.1002\/pro.70655","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/pro.70655","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,10]],"date-time":"2026-08-10T05:25:40Z","timestamp":1786339540000},"score":1,"resource":{"primary":{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/10.1002\/pro.70655"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,6,7]]},"references-count":29,"journal-issue":{"issue":"7","published-print":{"date-parts":[[2026,7]]}},"alternative-id":["10.1002\/pro.70655"],"URL":"https:\/\/doi.org\/10.1002\/pro.70655","archive":["Portico"],"relation":{},"ISSN":["0961-8368","1469-896X"],"issn-type":[{"value":"0961-8368","type":"print"},{"value":"1469-896X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,6,7]]},"assertion":[{"value":"2026-01-02","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-05-15","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-06-07","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"e70655"}}