{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"institution":[{"name":"bioRxiv"}],"indexed":{"date-parts":[[2026,9,16]],"date-time":"2026-09-16T11:46:54Z","timestamp":1789559214559,"version":"build-2803163510"},"posted":{"date-parts":[[2026,6,13]]},"group-title":"Bioinformatics","reference-count":17,"publisher":"openRxiv","license":[{"start":{"date-parts":[[2026,6,13]],"date-time":"2026-06-13T00:00:00Z","timestamp":1781308800000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"award":["R01-HG006677"],"award-info":[{"award-number":["R01-HG006677"]}],"id":[{"id":"https:\/\/ror.org\/00baak391","id-type":"ROR","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"accepted":{"date-parts":[[2026,6,13]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                <jats:p>Although AlphaFold2 and its competitors have demonstrated remarkable abilities to predict protein structure, more work is needed to explore the limitations of these methods. Here we investigated the reliability of AlphaFold2 and ColabFold by creating a set of realistic but false protein sequences, using ColabFold to predict their structure, and then asking how often the program produces a high-scoring structure for a sequence that does not represent a protein. We determined that AlphaFold2 has a very small but non-zero false positive rate, estimated here at approximately 1 in 435 if one uses a threshold pLDDT score of 70 to define positive predictions. We also discovered, serendipitously, that some high-scoring sequences in the human genome were not false positives, but instead were previously unknown and un-annotated pseudogenes. These latter findings indicate that some well-established human annotations of protein-coding genes may have incorrectly extended the 5\u2019 untranslated regions too far. They also suggest that AlphaFold2\u2019s false positive rate is low enough that almost any high-scoring structure, even in a noncoding region, is worthy of further investigation.<\/jats:p>","DOI":"10.64898\/2026.06.11.731682","type":"posted-content","created":{"date-parts":[[2026,6,14]],"date-time":"2026-06-14T05:25:11Z","timestamp":1781414711000},"source":"Crossref","is-referenced-by-count":1,"title":["Testing the reliability of AI-generated protein structures"],"prefix":"10.64898","author":[{"given":"Amanda","family":"Xu","sequence":"first","affiliation":[{"name":"Center for Computational Biology, Johns Hopkins University, Baltimore, Maryland"},{"name":"Department of Computer Science, Johns Hopkins University, Baltimore, Maryland"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8859-7432","authenticated-orcid":false,"given":"Steven L.","family":"Salzberg","sequence":"additional","affiliation":[{"name":"Center for Computational Biology, Johns Hopkins University, Baltimore, Maryland"},{"name":"Department of Computer Science, Johns Hopkins University, Baltimore, Maryland"},{"name":"Department of Biomedical Engineering, Johns Hopkins University, Baltimore, Maryland"},{"name":"Department of Biostatistics, Johns Hopkins University, Baltimore, Maryland"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"54368","reference":[{"key":"2026061620351171000_2026.06.11.731682v1.1","doi-asserted-by":"publisher","DOI":"10.1038\/S41586-021-03819-2"},{"issue":"6574","key":"2026061620351171000_2026.06.11.731682v1.2","doi-asserted-by":"crossref","first-page":"1426","DOI":"10.1126\/science.acz9822","article-title":"Protein structures for all","volume":"374","year":"2021","journal-title":"Science"},{"key":"2026061620351171000_2026.06.11.731682v1.3","doi-asserted-by":"publisher","DOI":"10.1093\/NAR\/GKAD1011"},{"key":"2026061620351171000_2026.06.11.731682v1.4","doi-asserted-by":"publisher","DOI":"10.1021\/acs.jcim.4c00976"},{"key":"2026061620351171000_2026.06.11.731682v1.5","doi-asserted-by":"crossref","unstructured":"Sommer, M.J. , S. Cha , A. Varabyou , N. Rincon , S. Park , I. Minkin , M. Pertea , M. Steinegger , and S.L. Salzberg . Structure-guided isoform identification for the human transcriptome. Elife, 2022, 11.","DOI":"10.7554\/eLife.82556"},{"key":"2026061620351171000_2026.06.11.731682v1.6","doi-asserted-by":"crossref","unstructured":"Monzon, V. , D.H. Haft , and A. Bateman . Folding the unfoldable: using AlphaFold to explore spurious proteins. Bioinformatics Advances, 2022, 2(1).","DOI":"10.1093\/bioadv\/vbab043"},{"key":"2026061620351171000_2026.06.11.731682v1.7","doi-asserted-by":"publisher","DOI":"10.1093\/database\/bas003"},{"key":"2026061620351171000_2026.06.11.731682v1.8","doi-asserted-by":"publisher","DOI":"10.1016\/j.csbj.2025.01.016"},{"key":"2026061620351171000_2026.06.11.731682v1.9","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-022-04558-8"},{"key":"2026061620351171000_2026.06.11.731682v1.10","doi-asserted-by":"publisher","DOI":"10.1038\/S41592-022-01488-1"},{"key":"2026061620351171000_2026.06.11.731682v1.11","doi-asserted-by":"publisher","DOI":"10.1186\/s13059-023-03088-4"},{"key":"2026061620351171000_2026.06.11.731682v1.12","doi-asserted-by":"publisher","DOI":"10.1093\/nar\/gkae1038"},{"key":"2026061620351171000_2026.06.11.731682v1.13","doi-asserted-by":"publisher","DOI":"10.1093\/nar\/gkae1078"},{"key":"2026061620351171000_2026.06.11.731682v1.14","doi-asserted-by":"publisher","DOI":"10.1093\/nar\/25.17.3389"},{"key":"2026061620351171000_2026.06.11.731682v1.15","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-023-06947-z"},{"issue":"1","key":"2026061620351171000_2026.06.11.731682v1.16","doi-asserted-by":"crossref","first-page":"100","DOI":"10.1038\/s42003-024-05779-5","article-title":"Transcription readthrough is prevalent in healthy human tissues and associated with inherent genomic features","volume":"7","year":"2024","journal-title":"Commun Biol"},{"key":"2026061620351171000_2026.06.11.731682v1.17","unstructured":"The PyMOL Molecular Graphics System, Version 2.5.2. 2023, Schr\u00f6dinger, LLC."}],"container-title":[],"original-title":[],"link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.64898\/2026.06.11.731682","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,17]],"date-time":"2026-06-17T03:35:20Z","timestamp":1781667320000},"score":1,"resource":{"primary":{"URL":"http:\/\/biorxiv.org\/lookup\/doi\/10.64898\/2026.06.11.731682"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,6,13]]},"references-count":17,"URL":"https:\/\/doi.org\/10.64898\/2026.06.11.731682","relation":{},"subject":[],"published":{"date-parts":[[2026,6,13]]},"subtype":"preprint"}}