{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,8]],"date-time":"2026-07-08T04:04:55Z","timestamp":1783483495822,"version":"3.55.0"},"reference-count":33,"publisher":"Oxford University Press (OUP)","issue":"2","license":[{"start":{"date-parts":[[2024,2,3]],"date-time":"2024-02-03T00:00:00Z","timestamp":1706918400000},"content-version":"vor","delay-in-days":2,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100000266","name":"Engineering and Physical Sciences Research Council","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100000266","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100000266","name":"EPSRC","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100000266","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Artificial and Augmented Intelligence for Automated Scientific Discovery","award":["EP\/S000356\/1"],"award-info":[{"award-number":["EP\/S000356\/1"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2024,2,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:sec>\n                  <jats:title>Motivation<\/jats:title>\n                  <jats:p>Protein language models (PLMs), which borrowed ideas for modelling and inference from natural language processing, have demonstrated the ability to extract meaningful representations in an unsupervised way. This led to significant performance improvement in several downstream tasks. Clustering amino acids based on their physical\u2013chemical properties to achieve reduced alphabets has been of interest in past research, but their application to PLMs or folding models is unexplored.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Results<\/jats:title>\n                  <jats:p>Here, we investigate the efficacy of PLMs trained on reduced amino acid alphabets in capturing evolutionary information, and we explore how the loss of protein sequence information impacts learned representations and downstream task performance. Our empirical work shows that PLMs trained on the full alphabet and a large number of sequences capture fine details that are lost in alphabet reduction methods. We further show the ability of a structure prediction model(ESMFold) to fold CASP14 protein sequences translated using a reduced alphabet. For 10 proteins out of the 50 targets, reduced alphabets improve structural predictions with LDDT-C\u03b1 differences of up to 19%.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Availability and implementation<\/jats:title>\n                  <jats:p>Trained models and code are available at github.com\/Ieremie\/reduced-alph-PLM.<\/jats:p>\n               <\/jats:sec>","DOI":"10.1093\/bioinformatics\/btae061","type":"journal-article","created":{"date-parts":[[2024,2,4]],"date-time":"2024-02-04T05:14:53Z","timestamp":1707023693000},"source":"Crossref","is-referenced-by-count":13,"title":["Protein language models meet reduced amino acid alphabets"],"prefix":"10.1093","volume":"40","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-2897-2541","authenticated-orcid":false,"given":"Ioan","family":"Ieremie","sequence":"first","affiliation":[{"name":"Vision, Learning & Control Group, University of Southampton , Southampton SO17 1BJ, United Kingdom"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Rob M","family":"Ewing","sequence":"additional","affiliation":[{"name":"Biological Sciences, University of Southampton , Southampton SO17 1BJ, United Kingdom"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mahesan","family":"Niranjan","sequence":"additional","affiliation":[{"name":"Vision, Learning & Control Group, University of Southampton , Southampton SO17 1BJ, United Kingdom"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"286","published-online":{"date-parts":[[2024,2,3]]},"reference":[{"key":"2024021622465068700_btae061-B1","doi-asserted-by":"crossref","first-page":"1315","DOI":"10.1038\/s41592-019-0598-1","article-title":"Unified rational protein engineering with sequence-based deep representation learning","volume":"16","author":"Alley","year":"2019","journal-title":"Nat Methods"},{"key":"2024021622465068700_btae061-B2","author":"Bepler","year":"2018"},{"key":"2024021622465068700_btae061-B3","doi-asserted-by":"crossref","first-page":"691","DOI":"10.1038\/s41587-020-00793-4","article-title":"Deep diversification of an AAV capsid protein by machine learning","volume":"39","author":"Bryant","year":"2021","journal-title":"Nat Biotechnol"},{"key":"2024021622465068700_btae061-B4","doi-asserted-by":"crossref","first-page":"59","DOI":"10.1038\/nmeth.3176","article-title":"Fast and sensitive protein alignment using diamond","volume":"12","author":"Buchfink","year":"2015","journal-title":"Nat Methods"},{"key":"2024021622465068700_btae061-B5","author":"Dallago","year":"2021"},{"key":"2024021622465068700_btae061-B6","doi-asserted-by":"crossref","first-page":"D482","DOI":"10.1093\/nar\/gky1114","article-title":"Sifts: updated structure integration with function, taxonomy and sequences resource allows 40-fold increase in coverage of structure-based annotations for proteins","volume":"47","author":"Dana","year":"2019","journal-title":"Nucleic Acids Res"},{"key":"2024021622465068700_btae061-B7","first-page":"4171","author":"Devlin","year":"2019"},{"key":"2024021622465068700_btae061-B8","doi-asserted-by":"crossref","first-page":"3751","DOI":"10.1073\/pnas.89.9.3751","article-title":"Folding and function of a t4 lysozyme containing 10 consecutive alanines illustrate the redundancy of information in an amino acid sequence","volume":"89","author":"Heinz","year":"1992","journal-title":"Proc Natl Acad Sci USA"},{"key":"2024021622465068700_btae061-B9","first-page":"1","author":"Hermosilla","year":"2021"},{"key":"2024021622465068700_btae061-B10","doi-asserted-by":"crossref","first-page":"495","DOI":"10.1038\/s41592-020-0801-4","article-title":"Meltome atlas\u2014thermal proteome stability across the tree of life","volume":"17","author":"Jarzab","year":"2020","journal-title":"Nat Methods"},{"key":"2024021622465068700_btae061-B11","doi-asserted-by":"crossref","first-page":"1711","DOI":"10.1002\/prot.26257","article-title":"Applying and improving alphafold at casp14","volume":"89","author":"Jumper","year":"2021","journal-title":"Proteins Struct Funct Bioinf"},{"key":"2024021622465068700_btae061-B12","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":"2024021622465068700_btae061-B13","doi-asserted-by":"crossref","first-page":"1607","DOI":"10.1002\/prot.26237","article-title":"Critical assessment of methods of protein structure prediction (CASP)\u2014round XIV","volume":"89","author":"Kryshtafovych","year":"2021","journal-title":"Proteins Struct Funct Bioinf"},{"key":"2024021622465068700_btae061-B14","doi-asserted-by":"crossref","first-page":"666","DOI":"10.1126\/science.273.5275.666","article-title":"Emergence of preferred structures in a simple model of protein folding","volume":"273","author":"Li","year":"1996","journal-title":"Science"},{"key":"2024021622465068700_btae061-B15","doi-asserted-by":"crossref","first-page":"3503","DOI":"10.1016\/j.csbj.2022.07.001","article-title":"Research progress of reduced amino acid alphabets in protein analysis and prediction","volume":"20","author":"Liang","year":"2022","journal-title":"Comput Struct Biotechnol J"},{"key":"2024021622465068700_btae061-B16","author":"Lin","year":"2022"},{"key":"2024021622465068700_btae061-B17","doi-asserted-by":"crossref","first-page":"2722","DOI":"10.1093\/bioinformatics\/btt473","article-title":"lddt: a local superposition-free score for comparing protein structures and models using distance difference tests","volume":"29","author":"Mariani","year":"2013","journal-title":"Bioinformatics"},{"key":"2024021622465068700_btae061-B18","doi-asserted-by":"crossref","first-page":"679","DOI":"10.1038\/s41592-022-01488-1","article-title":"Colabfold: making protein folding accessible to all","volume":"19","author":"Mirdita","year":"2022","journal-title":"Nat Methods"},{"key":"2024021622465068700_btae061-B19","doi-asserted-by":"crossref","first-page":"149","DOI":"10.1093\/protein\/13.3.149","article-title":"Simplified amino acid alphabets for protein fold recognition and implications for folding","volume":"13","author":"Murphy","year":"2000","journal-title":"Protein Eng"},{"key":"2024021622465068700_btae061-B20","first-page":"1","author":"Nambiar","year":"2020"},{"key":"2024021622465068700_btae061-B21","article-title":"Pytorch: an imperative style, high-performance deep learning library","volume":"32","author":"Paszke","year":"2019","journal-title":"Adv Neural Inf Process Syst"},{"key":"2024021622465068700_btae061-B22","article-title":"Evaluating protein transfer learning with tape","volume":"32","author":"Rao","year":"2019","journal-title":"Adv Neural Inf Process Syst"},{"key":"2024021622465068700_btae061-B23","doi-asserted-by":"crossref","first-page":"805","DOI":"10.1038\/nsb1097-805","article-title":"Functional rapidly folding proteins from simplified amino acid sequences","volume":"4","author":"Riddle","year":"1997","journal-title":"Nat Struct Biol"},{"key":"2024021622465068700_btae061-B24","doi-asserted-by":"crossref","first-page":"e2016239118","DOI":"10.1073\/pnas.2016239118","article-title":"Biological structure and function emerge from scaling unsupervised learning to 250 million protein sequences","volume":"118","author":"Rives","year":"2021","journal-title":"Proc Natl Acad Sci USA"},{"key":"2024021622465068700_btae061-B25","doi-asserted-by":"crossref","first-page":"149","DOI":"10.1002\/(SICI)1097-0134(20000201)38:2<149::AID-PROT4>3.0.CO;2-#","article-title":"Optimized representations and maximal information in proteins","volume":"38","author":"Solis","year":"2000","journal-title":"Proteins"},{"key":"2024021622465068700_btae061-B26","doi-asserted-by":"crossref","first-page":"1026","DOI":"10.1038\/nbt.3988","article-title":"Mmseqs2 enables sensitive protein sequence searching for the analysis of massive data sets","volume":"35","author":"Steinegger","year":"2017","journal-title":"Nat Biotechnol"},{"key":"2024021622465068700_btae061-B27","doi-asserted-by":"crossref","first-page":"2542","DOI":"10.1038\/s41467-018-04964-5","article-title":"Clustering huge protein sequence sets in linear time","volume":"9","author":"Steinegger","year":"2018","journal-title":"Nat Commun"},{"key":"2024021622465068700_btae061-B28","doi-asserted-by":"crossref","first-page":"1282","DOI":"10.1093\/bioinformatics\/btm098","article-title":"Uniref: comprehensive and non-redundant uniprot reference clusters","volume":"23","author":"Suzek","year":"2007","journal-title":"Bioinformatics"},{"key":"2024021622465068700_btae061-B29","doi-asserted-by":"crossref","first-page":"205","DOI":"10.1016\/S0022-5193(86)80075-3","article-title":"The classification of amino acid conservation","volume":"119","author":"Taylor","year":"1986","journal-title":"J Theor Biol"},{"key":"2024021622465068700_btae061-B30","doi-asserted-by":"crossref","first-page":"1033","DOI":"10.1038\/14918","article-title":"A computational approach to simplifying the protein folding alphabet","volume":"6","author":"Wang","year":"1999","journal-title":"Nat Struct Biol"},{"key":"2024021622465068700_btae061-B31","doi-asserted-by":"crossref","first-page":"e16965","DOI":"10.7554\/eLife.16965","article-title":"Adaptation in protein fitness landscapes is facilitated by indirect paths","volume":"5","author":"Wu","year":"2016","journal-title":"Elife"},{"key":"2024021622465068700_btae061-B32","author":"Yang","year":"2022"},{"key":"2024021622465068700_btae061-B33","author":"Zhang","year":"2022"}],"container-title":["Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/bioinformatics\/advance-article-pdf\/doi\/10.1093\/bioinformatics\/btae061\/56580314\/btae061.pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/40\/2\/btae061\/56675274\/btae061.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/40\/2\/btae061\/56675274\/btae061.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,2,16]],"date-time":"2024-02-16T22:48:37Z","timestamp":1708123717000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article\/doi\/10.1093\/bioinformatics\/btae061\/7600424"}},"subtitle":[],"editor":[{"given":"Xin","family":"Gao","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2024,2,1]]},"references-count":33,"journal-issue":{"issue":"2","published-print":{"date-parts":[[2024,2,1]]}},"URL":"https:\/\/doi.org\/10.1093\/bioinformatics\/btae061","relation":{},"ISSN":["1367-4803","1367-4811"],"issn-type":[{"value":"1367-4803","type":"print"},{"value":"1367-4811","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2024,2,1]]},"published":{"date-parts":[[2024,2,1]]},"article-number":"btae061"}}