{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,18]],"date-time":"2026-05-18T05:16:28Z","timestamp":1779081388750,"version":"3.51.4"},"reference-count":54,"publisher":"Oxford University Press (OUP)","issue":"18","license":[{"start":{"date-parts":[[2017,4,18]],"date-time":"2017-04-18T00:00:00Z","timestamp":1492473600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/academic.oup.com\/journals\/pages\/about_us\/legal\/notices"}],"funder":[{"DOI":"10.13039\/501100000925","name":"National Health and Medical Research Council","doi-asserted-by":"publisher","award":["1059775"],"award-info":[{"award-number":["1059775"]}],"id":[{"id":"10.13039\/501100000925","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100000925","name":"National Health and Medical Research Council","doi-asserted-by":"publisher","award":["1083450"],"award-info":[{"award-number":["1083450"]}],"id":[{"id":"10.13039\/501100000925","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100000923","name":"Australian Research Council","doi-asserted-by":"publisher","award":["LE150100161"],"award-info":[{"award-number":["LE150100161"]}],"id":[{"id":"10.13039\/501100000923","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2017,9,15]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:sec>\n                  <jats:title>Motivation<\/jats:title>\n                  <jats:p>The accuracy of predicting protein local and global structural properties such as secondary structure and solvent accessible surface area has been stagnant for many years because of the challenge of accounting for non-local interactions between amino acid residues that are close in three-dimensional structural space but far from each other in their sequence positions. All existing machine-learning techniques relied on a sliding window of 10\u201320 amino acid residues to capture some \u2018short to intermediate\u2019 non-local interactions. Here, we employed Long Short-Term Memory (LSTM) Bidirectional Recurrent Neural Networks (BRNNs) which are capable of capturing long range interactions without using a window.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Results<\/jats:title>\n                  <jats:p>We showed that the application of LSTM-BRNN to the prediction of protein structural properties makes the most significant improvement for residues with the most long-range contacts (|i-j|\u2009&amp;gt;19) over a previous window-based, deep-learning method SPIDER2. Capturing long-range interactions allows the accuracy of three-state secondary structure prediction to reach 84% and the correlation coefficient between predicted and actual solvent accessible surface areas to reach 0.80, plus a reduction of 5%, 10%, 5% and 10% in the mean absolute error for backbone \u03d5, \u03c8, \u03b8 and \u03c4 angles, respectively, from SPIDER2. More significantly, 27% of 182724\u200940-residue models directly constructed from predicted C\u03b1 atom-based \u03b8 and \u03c4 have similar structures to their corresponding native structures (6\u00c5 RMSD or less), which is 3% better than models built by \u03d5 and \u03c8 angles. We expect the method to be useful for assisting protein structure and function prediction.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Availability and implementation<\/jats:title>\n                  <jats:p>The method is available as a SPIDER3 server and standalone package at http:\/\/sparks-lab.org.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Supplementary information<\/jats:title>\n                  <jats:p>Supplementary data are available at Bioinformatics online.<\/jats:p>\n               <\/jats:sec>","DOI":"10.1093\/bioinformatics\/btx218","type":"journal-article","created":{"date-parts":[[2017,4,15]],"date-time":"2017-04-15T11:07:38Z","timestamp":1492254458000},"page":"2842-2849","source":"Crossref","is-referenced-by-count":348,"title":["Capturing non-local interactions by long short-term memory bidirectional recurrent neural networks for improving prediction of protein secondary structure, backbone angles, contact numbers and solvent accessibility"],"prefix":"10.1093","volume":"33","author":[{"given":"Rhys","family":"Heffernan","sequence":"first","affiliation":[{"name":"Signal Processing Laboratory, Griffith University, Brisbane, QLD, Australia"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yuedong","family":"Yang","sequence":"additional","affiliation":[{"name":"Institute for Glycomics and School of Information and Communication Technology, Griffith University, Southport, QLD, Australia"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Kuldip","family":"Paliwal","sequence":"additional","affiliation":[{"name":"Signal Processing Laboratory, Griffith University, Brisbane, QLD, Australia"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yaoqi","family":"Zhou","sequence":"additional","affiliation":[{"name":"Institute for Glycomics and School of Information and Communication Technology, Griffith University, Southport, QLD, Australia"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"286","published-online":{"date-parts":[[2017,4,18]]},"reference":[{"key":"2023020206412600100_btx218-B1","author":"Abadi","year":"2016"},{"key":"2023020206412600100_btx218-B2","doi-asserted-by":"crossref","first-page":"753","DOI":"10.1002\/prot.20176","article-title":"Accurate prediction of solvent accessibility using neural networks-based regression","volume":"56","author":"Adamczak","year":"2004","journal-title":"Proteins"},{"key":"2023020206412600100_btx218-B3","doi-asserted-by":"crossref","first-page":"629","DOI":"10.1002\/prot.10328","article-title":"Real value prediction of solvent accessibility from amino acid sequence","volume":"50","author":"Ahmad","year":"2003","journal-title":"Proteins"},{"key":"2023020206412600100_btx218-B4","doi-asserted-by":"crossref","first-page":"3389","DOI":"10.1093\/nar\/25.17.3389","article-title":"Gapped BLAST and PSI-BLAST: a new generation of protein database search programs","volume":"25","author":"Altschul","year":"1997","journal-title":"Nucleic Acids Res"},{"key":"2023020206412600100_btx218-B5","author":"Amodei","year":"2015"},{"key":"2023020206412600100_btx218-B6","doi-asserted-by":"crossref","first-page":"937","DOI":"10.1093\/bioinformatics\/15.11.937","article-title":"Exploiting the past and the future in protein secondary structure prediction","volume":"15","author":"Baldi","year":"1999","journal-title":"Bioinformatics"},{"key":"2023020206412600100_btx218-B7","doi-asserted-by":"crossref","first-page":"1042","DOI":"10.1126\/science.1219021","article-title":"The protein-folding problem, 50\u2009years on","volume":"338","author":"Dill","year":"2012","journal-title":"Science"},{"key":"2023020206412600100_btx218-B8","doi-asserted-by":"crossref","first-page":"76","DOI":"10.1002\/prot.21408","article-title":"Real-SPINE: an integrated system of neural networks for real-value prediction of protein structural properties","volume":"68","author":"Dor","year":"2007","journal-title":"Protein"},{"key":"2023020206412600100_btx218-B9","doi-asserted-by":"crossref","first-page":"W389","DOI":"10.1093\/nar\/gkv332","article-title":"JPred4: a protein secondary structure prediction server","volume":"43","author":"Drozdetskiy","year":"2015","journal-title":"Nucleic Acids Res"},{"key":"2023020206412600100_btx218-B10","doi-asserted-by":"crossref","first-page":"259","DOI":"10.1002\/jcc.21968","article-title":"SPINE X: improving protein secondary structure prediction by multistep learning coupled with prediction of solvent accessible surface area and backbone torsion angles","volume":"33","author":"Faraggi","year":"2012","journal-title":"J. Comput. Chem"},{"key":"2023020206412600100_btx218-B11","doi-asserted-by":"crossref","first-page":"269","DOI":"10.1111\/j.1399-3011.1988.tb01261.x","article-title":"Amino acid side chain parameters for correlation studies in biology and pharmacology","volume":"32","author":"Fauch\u00e8re","year":"1988","journal-title":"Int. J. Pept. Protein Res"},{"key":"2023020206412600100_btx218-B12","doi-asserted-by":"crossref","first-page":"318","DOI":"10.1002\/prot.20630","article-title":"Real value prediction of solvent accessibility in proteins using multiple sequence alignment and secondary structure","volume":"61","author":"Garg","year":"2005","journal-title":"Proteins"},{"key":"2023020206412600100_btx218-B13","doi-asserted-by":"crossref","first-page":"420","DOI":"10.1073\/pnas.58.2.420","article-title":"Minimization of polypeptide energy. i. preliminary structures of bovine pancreatic ribonuclease s-peptide","volume":"58","author":"Gibson","year":"1967","journal-title":"Proc. Natl. Acad. Sci. USA"},{"key":"2023020206412600100_btx218-B14","doi-asserted-by":"crossref","first-page":"276","DOI":"10.1006\/jmbi.1997.1237","article-title":"Predicting protein stability changes upon mutation using database-derived potentials: solvent accessibility determines the importance of local versus non-local interactions along the sequence","volume":"272","author":"Gilis","year":"1997","journal-title":"J. Mol. Biol"},{"key":"2023020206412600100_btx218-B15","first-page":"545","volume-title":"Advances in Neural Information Processing Systems 21","author":"Graves","year":"2009"},{"key":"2023020206412600100_btx218-B16","doi-asserted-by":"crossref","first-page":"38","DOI":"10.1002\/prot.20379","article-title":"An amino acid has two sides: a new 2D measure provides a different view of solvent exposure","volume":"59","author":"Hamelryck","year":"2005","journal-title":"Proteins"},{"key":"2023020206412600100_btx218-B17","doi-asserted-by":"crossref","DOI":"10.1093\/bioinformatics\/btw678","article-title":"Improving protein disorder prediction by deep bidirectional long short-term memory recurrent neural networks","author":"Hanson","year":"2017","journal-title":"Bioinformatics"},{"key":"2023020206412600100_btx218-B18","doi-asserted-by":"crossref","first-page":"11476","DOI":"10.1038\/srep11476","article-title":"Improving prediction of secondary structure, local backbone angles, and solvent accessible surface area of proteins by iterative deep learning","volume":"5","author":"Heffernan","year":"2015","journal-title":"Sci. Rep"},{"key":"2023020206412600100_btx218-B19","doi-asserted-by":"crossref","first-page":"843","DOI":"10.1093\/bioinformatics\/btv665","article-title":"Highly accurate sequence-based prediction of half-sphere exposures of amino acid residues in proteins","volume":"32","author":"Heffernan","year":"2016","journal-title":"Bioinformatics"},{"key":"2023020206412600100_btx218-B20","doi-asserted-by":"crossref","first-page":"1735","DOI":"10.1162\/neco.1997.9.8.1735","article-title":"Long short-term memory","volume":"9","author":"Hochreiter","year":"1997","journal-title":"Neural Comput"},{"key":"2023020206412600100_btx218-B21","doi-asserted-by":"crossref","first-page":"659","DOI":"10.1093\/protein\/3.8.659","article-title":"Predicting surface exposure of amino acids from protein sequence","volume":"3","author":"Holbrook","year":"1990","journal-title":"Protein Eng"},{"key":"2023020206412600100_btx218-B22","doi-asserted-by":"crossref","first-page":"195","DOI":"10.1006\/jmbi.1999.3091","article-title":"Protein secondary structure prediction based on position-specific scoring matrices","volume":"292","author":"Jones","year":"1999","journal-title":"J. Mol. Biol"},{"key":"2023020206412600100_btx218-B23","doi-asserted-by":"crossref","first-page":"2577","DOI":"10.1002\/bip.360221211","article-title":"Dictionary of protein secondary structure: pattern recognition of hydrogen-bonded and geometrical features","volume":"22","author":"Kabsch","year":"1983","journal-title":"Biopolymers"},{"key":"2023020206412600100_btx218-B24","doi-asserted-by":"crossref","first-page":"448","DOI":"10.1006\/jmbi.1993.1045","article-title":"Estimation and use of protein backbone angle probabilities","volume":"229","author":"Kang","year":"1993","journal-title":"J Mol. Biol"},{"key":"2023020206412600100_btx218-B25","author":"Kingma","year":"2014"},{"key":"2023020206412600100_btx218-B26","doi-asserted-by":"crossref","first-page":"401","DOI":"10.1186\/1471-2105-7-401","article-title":"CRNPRED: highly accurate prediction of one-dimensional protein structures by large-scale critical random networks","volume":"7","author":"Kinjo","year":"2006","journal-title":"BMC Bioinformatics"},{"key":"2023020206412600100_btx218-B27","doi-asserted-by":"crossref","first-page":"1612","DOI":"10.1093\/bioinformatics\/bth136","article-title":"Protein backbone angle prediction with machine learning approaches","volume":"20","author":"Kuang","year":"2004","journal-title":"Bioinformatics"},{"key":"2023020206412600100_btx218-B28","volume-title":"Predicting Structured Data","author":"LeCun","year":"2006"},{"key":"2023020206412600100_btx218-B29","doi-asserted-by":"crossref","first-page":"379","DOI":"10.1016\/0022-2836(71)90324-X","article-title":"The interpretation of protein structures: estimation of static accessibility","volume":"55","author":"Lee","year":"1971","journal-title":"J. Mol. Biol"},{"key":"2023020206412600100_btx218-B30","doi-asserted-by":"crossref","first-page":"2040","DOI":"10.1002\/jcc.23718","article-title":"Predicting backbone c\u03b1 angles and dihedrals from protein sequences by stacked sparse auto-encoder deep neural network","volume":"35","author":"Lyons","year":"2014","journal-title":"J. Comput. Chem"},{"key":"2023020206412600100_btx218-B31","doi-asserted-by":"crossref","first-page":"2592","DOI":"10.1093\/bioinformatics\/btu352","article-title":"SSpro\/ACCpro 5: almost perfect prediction of protein secondary structure and relative solvent accessibility using profiles, machine learning and structural similarity","volume":"30","author":"Magnan","year":"2014","journal-title":"Bioinformatics"},{"key":"2023020206412600100_btx218-B32","doi-asserted-by":"crossref","first-page":"2056","DOI":"10.1093\/bioinformatics\/btt344","article-title":"Porter, PaleAle 4.0: high-accuracy prediction of protein secondary structure and relative solvent accessibility","volume":"29","author":"Mirabello","year":"2013","journal-title":"Bioinformatics"},{"key":"2023020206412600100_btx218-B33","doi-asserted-by":"crossref","first-page":"205","DOI":"10.1073\/pnas.37.4.205","article-title":"The structure of proteins: two hydrogen-bonded helical configurations of the polypeptide chain","volume":"37","author":"Pauling","year":"1951","journal-title":"Proc. Natl. Acad. Sci. USA"},{"key":"2023020206412600100_btx218-B34","doi-asserted-by":"crossref","first-page":"228","DOI":"10.1002\/prot.10082","article-title":"Improving the prediction of protein secondary structure in three and eight classes using recurrent neural networks and profiles","volume":"47","author":"Pollastri","year":"2002","journal-title":"Proteins"},{"key":"2023020206412600100_btx218-B35","doi-asserted-by":"crossref","first-page":"142","DOI":"10.1002\/prot.10069","article-title":"Prediction of coordination number and relative solvent accessibility in proteins","volume":"47","author":"Pollastri","year":"2002","journal-title":"Proteins"},{"key":"2023020206412600100_btx218-B36","doi-asserted-by":"crossref","first-page":"173","DOI":"10.1038\/nmeth.1818","article-title":"HHblits: lightning-fast iterative protein sequence searching by HMM-HMM alignment","volume":"9","author":"Remmert","year":"2012","journal-title":"Nat. Methods"},{"key":"2023020206412600100_btx218-B37","doi-asserted-by":"crossref","first-page":"141","DOI":"10.1016\/S1359-0278(98)00019-4","article-title":"What is the probability of a chance prediction of a protein structure with an rmsd of 6 \u00c5?","volume":"3","author":"Reva","year":"1998","journal-title":"Fold Des"},{"key":"2023020206412600100_btx218-B38","doi-asserted-by":"crossref","first-page":"204","DOI":"10.1006\/jsbi.2001.4336","article-title":"Review: protein secondary structure prediction continues to rise","volume":"134","author":"Rost","year":"2001","journal-title":"J. Struct. Biol"},{"key":"2023020206412600100_btx218-B39","doi-asserted-by":"crossref","first-page":"216","DOI":"10.1002\/prot.340200303","article-title":"Conservation and prediction of solvent accessibility in protein families","volume":"20","author":"Rost","year":"1994","journal-title":"Proteins"},{"key":"2023020206412600100_btx218-B40","doi-asserted-by":"crossref","first-page":"2673","DOI":"10.1109\/78.650093","article-title":"Bidirectional recurrent neural networks","volume":"45","author":"Schuster","year":"1997","journal-title":"IEEE Trans. Signal Process"},{"key":"2023020206412600100_btx218-B41","doi-asserted-by":"crossref","first-page":"1489","DOI":"10.1093\/bioinformatics\/btn222","article-title":"HSEpred: predict half-sphere exposure from protein sequences","volume":"24","author":"Song","year":"2008","journal-title":"Bioinformatics"},{"key":"2023020206412600100_btx218-B42","first-page":"1929","article-title":"Dropout: a simple way to prevent neural networks from overfitting","volume":"15","author":"Srivastava","year":"2014","journal-title":"J. Mach. Learn. Res"},{"key":"2023020206412600100_btx218-B43","author":"Sundermeyer","year":"2012"},{"key":"2023020206412600100_btx218-B44","doi-asserted-by":"crossref","first-page":"1513","DOI":"10.1093\/bioinformatics\/btp240","article-title":"Identification of computational hot spots in protein interfaces: combining solvent accessibility and inter-residue potentials improves the accuracy","volume":"25","author":"Tuncbag","year":"2009","journal-title":"Bioinformatics"},{"key":"2023020206412600100_btx218-B45","doi-asserted-by":"crossref","first-page":"18962","DOI":"10.1038\/srep18962","article-title":"Protein secondary structure prediction using deep convolutional neural fields","volume":"6","author":"Wang","year":"2016","journal-title":"Sci. Rep"},{"key":"2023020206412600100_btx218-B46","doi-asserted-by":"crossref","first-page":"476","DOI":"10.1002\/prot.20435","article-title":"Protein secondary structure prediction with dihedral angles","volume":"59","author":"Wood","year":"2005","journal-title":"Proteins"},{"key":"2023020206412600100_btx218-B47","author":"Xiong","year":"2016"},{"key":"2023020206412600100_btx218-B48","doi-asserted-by":"crossref","first-page":"427","DOI":"10.1002\/prot.21940","article-title":"Real-value prediction of backbone torsion angles","volume":"72","author":"Xue","year":"2008","journal-title":"Proteins"},{"key":"2023020206412600100_btx218-B49","doi-asserted-by":"crossref","DOI":"10.1093\/bib\/bbw129","article-title":"Sixty-five years of long march in protein secondary structure prediction: the final stretch?","author":"Yang","year":"2016","journal-title":"Brief. Bioinform"},{"key":"2023020206412600100_btx218-B50","doi-asserted-by":"crossref","first-page":"992","DOI":"10.1021\/ci400647u","article-title":"Context-based features enhance protein secondary structure prediction accuracy","volume":"54","author":"Yaseen","year":"2014","journal-title":"J. Chem. Inf. Model"},{"key":"2023020206412600100_btx218-B51","doi-asserted-by":"crossref","first-page":"248","DOI":"10.1186\/1471-2105-6-248","article-title":"Better prediction of protein contact number using a support vector regression analysis of amino acid sequence","volume":"6","author":"Yuan","year":"2005","journal-title":"BMC Bioinformatics"},{"key":"2023020206412600100_btx218-B52","doi-asserted-by":"crossref","first-page":"558","DOI":"10.1002\/prot.20234","article-title":"Prediction of protein accessible surface areas by support vector regression","volume":"57","author":"Yuan","year":"2004","journal-title":"Proteins"},{"key":"2023020206412600100_btx218-B53","doi-asserted-by":"crossref","first-page":"45","DOI":"10.1002\/9780470882207.ch4","volume-title":"Introduction to Protein Structure Prediction","author":"Zhou","year":"2010"},{"key":"2023020206412600100_btx218-B54","doi-asserted-by":"crossref","first-page":"3","DOI":"10.1007\/s00214-010-0799-2","article-title":"Trends in template\/fragment-free protein structure prediction","volume":"128","author":"Zhou","year":"2011","journal-title":"Theor. Chem. Acc"}],"container-title":["Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/33\/18\/2842\/49040985\/bioinformatics_33_18_2842.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/33\/18\/2842\/49040985\/bioinformatics_33_18_2842.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,2,2]],"date-time":"2023-02-02T06:41:58Z","timestamp":1675320118000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article\/33\/18\/2842\/3738544"}},"subtitle":[],"editor":[{"given":"Alfonso","family":"Valencia","sequence":"additional","affiliation":[],"role":[{"role":"editor","vocabulary":"crossref"}]}],"short-title":[],"issued":{"date-parts":[[2017,4,18]]},"references-count":54,"journal-issue":{"issue":"18","published-print":{"date-parts":[[2017,9,15]]}},"URL":"https:\/\/doi.org\/10.1093\/bioinformatics\/btx218","relation":{},"ISSN":["1367-4803","1367-4811"],"issn-type":[{"value":"1367-4803","type":"print"},{"value":"1367-4811","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2017,9,15]]},"published":{"date-parts":[[2017,4,18]]}}}