{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,11]],"date-time":"2026-09-11T14:09:14Z","timestamp":1789135754674,"version":"build-2803163510"},"reference-count":29,"publisher":"Oxford University Press (OUP)","issue":"1","license":[{"start":{"date-parts":[[2026,8,6]],"date-time":"2026-08-06T00:00:00Z","timestamp":1785974400000},"content-version":"vor","delay-in-days":217,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Genotoul bioinformatics platform Toulouse Occitanie"},{"name":"bioinformatics and genomics platform, BIG, Sophia Antipolis"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,1,7]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Motivation<\/jats:title>\n                    <jats:p>Transformer-based models (TBMs) are state-of-the-art deep learning architectures that predict protein structural features with high accuracy. Despite methodological differences, they all rely on large datasets structured in families of homologous sequences. However, 5%\u201330% of eukaryotic proteomes consist of orphan proteins, which are sequences without detectable similarity to known families. Although they may share structural traits with characterized proteins, their lack of homology makes them an ideal dataset for evaluating TBM generalization beyond familiar sequence space.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>We compared predictions from several widely used TBM architectures on an expert-curated set of orphan proteins from the Meloidogyne genus, comprising some of the most destructive plant-parasitic nematodes. Multiple sequence alignment\u2013based approaches such as AlphaFold2 performed poorly on orphan proteins, as did single-sequence or embedding-based language models ESMFold, OmegaFold, and ProtT5. This limited performance cannot be fully attributed to intrinsic disorder, as confirmed by independent non-TBM disorder predictors. While accurate tertiary structure prediction remains out of reach, secondary structure is more reliably captured: predictors share about 70% of secondary structure elements, regardless of global fold similarity, and these elements are consistently identified by dedicated secondary structure tools.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Availability<\/jats:title>\n                    <jats:p>All data and analysis scripts are available at https:\/\/doi.org\/10.5281\/zenodo.18788931<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1093\/bioadv\/vbag215","type":"journal-article","created":{"date-parts":[[2026,7,30]],"date-time":"2026-07-30T11:59:07Z","timestamp":1785412747000},"source":"Crossref","is-referenced-by-count":0,"title":["Evaluating transformer-based models for structural characterization of orphan proteins"],"prefix":"10.1093","volume":"6","author":[{"ORCID":"https:\/\/orcid.org\/0009-0003-6821-6967","authenticated-orcid":false,"given":"Ercan","family":"Se\u00e7kin","sequence":"first","affiliation":[{"name":"Institut Sophia Agrobiotech, INRAE, Universit\u00e9 C\u00f4te d\u2019Azur, CNRS , Sophia-Antipolis 06903,","place":["France"]},{"name":"Inria Center of Universit\u00e9 C\u00f4te d\u2019Azur , Sophia Antipolis 06902,","place":["France"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5122-5072","authenticated-orcid":false,"given":"Dominique","family":"Colinet","sequence":"additional","affiliation":[{"name":"Institut Sophia Agrobiotech, INRAE, Universit\u00e9 C\u00f4te d\u2019Azur, CNRS , Sophia-Antipolis 06903,","place":["France"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4146-5608","authenticated-orcid":false,"given":"Etienne G J","family":"Danchin","sequence":"additional","affiliation":[{"name":"Institut Sophia Agrobiotech, INRAE, Universit\u00e9 C\u00f4te d\u2019Azur, CNRS , Sophia-Antipolis 06903,","place":["France"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2417-1448","authenticated-orcid":false,"given":"Edoardo","family":"Sarti","sequence":"additional","affiliation":[{"name":"Inria Center of Universit\u00e9 C\u00f4te d\u2019Azur , Sophia Antipolis 06902,","place":["France"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"286","published-online":{"date-parts":[[2026,8,6]]},"reference":[{"key":"2026091109314885300_vbag215-B1","doi-asserted-by":"crossref","first-page":"1056","DOI":"10.1038\/s41594-022-00849-w","article-title":"A structural biology community assessment of AlphaFold2 applications","volume":"29","author":"Akdel","year":"2022","journal-title":"Nat Struct Mol Biol"},{"key":"2026091109314885300_vbag215-B2","first-page":"1510","article-title":"De novo gene emergence and the dark proteome","volume":"9","author":"Basile","year":"2017","journal-title":"Genome Biol Evol"},{"key":"2026091109314885300_vbag215-B3","doi-asserted-by":"crossref","first-page":"vbag186","DOI":"10.1093\/bioadv\/vbag186","article-title":"Characterizing the fragmentation of AlphaFold predictions","author":"Cazals","year":"2026","journal-title":"Bioinformatics Adv"},{"key":"2026091109314885300_vbag215-B4","first-page":"898","article-title":"Identifying abundant and purifying selection pressures in protein sequence evolution","volume":"2","author":"Ekman","year":"2010","journal-title":"Genome Biol Evol"},{"key":"2026091109314885300_vbag215-B5","doi-asserted-by":"crossref","first-page":"7112","DOI":"10.1109\/TPAMI.2021.3095381","article-title":"Prottrans: toward understanding the language of life through self-supervised learning","volume":"44","author":"Elnaggar","year":"2022","journal-title":"IEEE Trans Pattern Anal Mach Intell"},{"key":"2026091109314885300_vbag215-B6","doi-asserted-by":"crossref","first-page":"W176","DOI":"10.1093\/nar\/gkae385","article-title":"AIUPred: combining energy estimation with deep learning for the enhanced prediction of protein disorder","volume":"52","author":"Erd\u0151s","year":"2024","journal-title":"Nucleic Acids Res"},{"key":"2026091109314885300_vbag215-B7","doi-asserted-by":"crossref","first-page":"27","DOI":"10.3390\/jdb11020027","article-title":"The lost and found: unraveling the functions of orphan genes","volume":"11","author":"Fakhar","year":"2023","journal-title":"J Dev Biol"},{"key":"2026091109314885300_vbag215-B8","author":"Hassan","year":"2025"},{"key":"2026091109314885300_vbag215-B9","doi-asserted-by":"crossref","first-page":"382","DOI":"10.1007\/s00239-020-09939-z","article-title":"A continuum of evolving de novo genes drives protein-coding novelty in drosophila","volume":"88","author":"Heames","year":"2020","journal-title":"J Mol Evol"},{"key":"2026091109314885300_vbag215-B10","doi-asserted-by":"crossref","first-page":"lqae150","DOI":"10.1093\/nargab\/lqae150","article-title":"Bilingual language model for protein sequence and structure","volume":"6","author":"Heinzinger","year":"2024","journal-title":"NAR Genom Bioinform"},{"key":"2026091109314885300_vbag215-B11","doi-asserted-by":"crossref","first-page":"e70208","DOI":"10.1002\/pro.70208","article-title":"DSSP 4: FAIR annotation of protein secondary structure","volume":"34","author":"Hekkelman","year":"2025","journal-title":"Protein Sci"},{"key":"2026091109314885300_vbag215-B12","doi-asserted-by":"crossref","first-page":"4438","DOI":"10.1038\/s41467-021-24773-7","article-title":"flDPnn: Accurate intrinsic disorder prediction with putative propensities of disorder functions","volume":"12","author":"Hu","year":"2021","journal-title":"Nat Commun"},{"key":"2026091109314885300_vbag215-B13","doi-asserted-by":"crossref","first-page":"583","DOI":"10.1038\/s41586-021-03819-2","article-title":"Highly accurate protein structure prediction with alphafold","volume":"596","author":"Jumper","year":"2021","journal-title":"Nature"},{"key":"2026091109314885300_vbag215-B14","doi-asserted-by":"crossref","first-page":"1160","DOI":"10.1093\/bib\/bbx108","article-title":"MAFFT online service: multiple sequence alignment, interactive sequence choice and visualization","volume":"20","author":"Katoh","year":"2019","journal-title":"Brief Bioinform"},{"key":"2026091109314885300_vbag215-B15","doi-asserted-by":"crossref","first-page":"1123","DOI":"10.1126\/science.ade2574","article-title":"Evolutionary-scale prediction of atomic-level protein structure with a language model","volume":"379","author":"Lin","year":"2023","journal-title":"Science"},{"key":"2026091109314885300_vbag215-B16","doi-asserted-by":"crossref","first-page":"i439","DOI":"10.1093\/bioinformatics\/btaf185","article-title":"LoRA-DR-suite: adapted embeddings predict intrinsic and soft disorder from protein sequences","volume":"41","author":"Lombardi","year":"2025","journal-title":"Bioinformatics"},{"key":"2026091109314885300_vbag215-B17","doi-asserted-by":"crossref","first-page":"757","DOI":"10.1002\/prot.26652","article-title":"Random, de novo, and conserved proteins: how structure and disorder predictors perform differently","volume":"92","author":"Middendorf","year":"2024","journal-title":"Proteins"},{"key":"2026091109314885300_vbag215-B18","doi-asserted-by":"crossref","first-page":"e4466","DOI":"10.1002\/pro.4466","article-title":"Intrinsic protein disorder and conditional folding in alphafolddb","volume":"31","author":"Piovesan","year":"2022","journal-title":"Protein Sci"},{"key":"2026091109314885300_vbag215-B19","doi-asserted-by":"crossref","first-page":"467","DOI":"10.1016\/j.csbj.2025.01.016","article-title":"AlphaFold 2, but not AlphaFold 3, predicts confident but unrealistic beta-solenoid structures for repeat proteins","volume":"27","author":"Pratt","year":"2025","journal-title":"Comput Struct Biotechnol J"},{"key":"2026091109314885300_vbag215-B20","doi-asserted-by":"crossref","first-page":"vbag190","DOI":"10.1093\/bioadv\/vbag190","article-title":"Fold or flop: quality assessment of AlphaFold predictions on whole proteomes","volume":"6","author":"Sarti","year":"2026","journal-title":"Bioinform Adv"},{"key":"2026091109314885300_vbag215-B21","first-page":"2906","article-title":"Insights into orphan gene evolution and disorder content","volume":"10","author":"Schmitz","year":"2018","journal-title":"Genome Biol Evol"},{"key":"2026091109314885300_vbag215-B22","author":"Se\u00e7kin","year":"2025"},{"key":"2026091109314885300_vbag215-B23","doi-asserted-by":"crossref","first-page":"evaf220","DOI":"10.1093\/gbe\/evaf220","article-title":"Orphan and de novo genes in fungi and animals: identification, origins and functions","volume":"17","author":"Se\u00e7kin","year":"2025","journal-title":"Genome Biol Evol"},{"key":"2026091109314885300_vbag215-B24","doi-asserted-by":"crossref","first-page":"243","DOI":"10.1038\/s41587-023-01773-0","article-title":"Fast and accurate protein structure search with foldseek","volume":"42","author":"van Kempen","year":"2024","journal-title":"Nat Biotechnol"},{"key":"2026091109314885300_vbag215-B25","author":"Williams","year":"2025"},{"key":"2026091109314885300_vbag215-B26","doi-asserted-by":"crossref","first-page":"0146","DOI":"10.1038\/s41559-017-0146","article-title":"Young genes are highly disordered as predicted by the preadaptation hypothesis of de novo gene birth","volume":"1","author":"Wilson","year":"2017","journal-title":"Nat Ecol Evol"},{"key":"2026091109314885300_vbag215-B27","author":"Wu","year":"2022"},{"key":"2026091109314885300_vbag215-B28","doi-asserted-by":"crossref","first-page":"2302","DOI":"10.1093\/nar\/gki524","article-title":"TM-align: a protein structure alignment algorithm based on the TM-score","volume":"33","author":"Zhang","year":"2005","journal-title":"Nucleic Acids Res"},{"key":"2026091109314885300_vbag215-B29","doi-asserted-by":"crossref","first-page":"e2406285121","DOI":"10.1073\/pnas.2406285121","article-title":"Protein language models learn evolutionary statistics of interacting sequence motifs","volume":"121","author":"Zhang","year":"2024","journal-title":"Proc Natl Acad Sci USA"}],"container-title":["Bioinformatics Advances"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/bioinformaticsadvances\/advance-article-pdf\/doi\/10.1093\/bioadv\/vbag215\/70510249\/vbag215.pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bioinformaticsadvances\/article-pdf\/6\/1\/vbag215\/70510249\/vbag215.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bioinformaticsadvances\/article-pdf\/6\/1\/vbag215\/70510249\/vbag215.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,11]],"date-time":"2026-09-11T13:32:02Z","timestamp":1789133522000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bioinformaticsadvances\/article\/doi\/10.1093\/bioadv\/vbag215\/8753557"}},"subtitle":[],"editor":[{"given":"Michael","family":"Gromiha","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026]]},"references-count":29,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2026,1,7]]}},"URL":"https:\/\/doi.org\/10.1093\/bioadv\/vbag215","relation":{},"ISSN":["2635-0041"],"issn-type":[{"value":"2635-0041","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2026]]},"published":{"date-parts":[[2026]]},"article-number":"vbag215"}}