{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,9]],"date-time":"2026-04-09T10:32:49Z","timestamp":1775730769935,"version":"3.50.1"},"reference-count":57,"publisher":"Oxford University Press (OUP)","issue":"23","license":[{"start":{"date-parts":[[2018,6,20]],"date-time":"2018-06-20T00:00:00Z","timestamp":1529452800000},"content-version":"vor","delay-in-days":1,"URL":"https:\/\/academic.oup.com\/journals\/pages\/open_access\/funder_policies\/chorus\/standard_publication_model"}],"funder":[{"DOI":"10.13039\/501100000923","name":"Australia Research Council","doi-asserted-by":"crossref","award":["DP180102060"],"award-info":[{"award-number":["DP180102060"]}],"id":[{"id":"10.13039\/501100000923","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/501100000925","name":"National Health and Medical Research Council","doi-asserted-by":"publisher","award":["1121629"],"award-info":[{"award-number":["1121629"]}],"id":[{"id":"10.13039\/501100000925","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["U1611261"],"award-info":[{"award-number":["U1611261"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["61772566"],"award-info":[{"award-number":["61772566"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Guangdong Introducing Innovative and Entrepreneurial Teams","award":["2016ZT06D211"],"award-info":[{"award-number":["2016ZT06D211"]}]},{"name":"High Performance Computing Cluster \u2018Gowonda\u2019"},{"DOI":"10.13039\/100015449","name":"Queensland Cyber Infrastructure Foundation","doi-asserted-by":"crossref","id":[{"id":"10.13039\/100015449","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/100015449","name":"QCIF","doi-asserted-by":"crossref","id":[{"id":"10.13039\/100015449","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2018,12,1]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:sec><jats:title>Motivation<\/jats:title><jats:p>Accurate prediction of a protein contact map depends greatly on capturing as much contextual information as possible from surrounding residues for a target residue pair. Recently, ultra-deep residual convolutional networks were found to be state-of-the-art in the latest Critical Assessment of Structure Prediction techniques (CASP12) for protein contact map prediction by attempting to provide a protein-wide context at each residue pair. Recurrent neural networks have seen great success in recent protein residue classification problems due to their ability to propagate information through long protein sequences, especially Long Short-Term Memory (LSTM) cells. Here, we propose a novel protein contact map prediction method by stacking residual convolutional networks with two-dimensional residual bidirectional recurrent LSTM networks, and using both one-dimensional sequence-based and two-dimensional evolutionary coupling-based information.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>We show that the proposed method achieves a robust performance over validation and independent test sets with the Area Under the receiver operating characteristic Curve (AUC) &amp;gt; 0.95 in all tests. When compared to several state-of-the-art methods for independent testing of 228 proteins, the method yields an AUC value of 0.958, whereas the next-best method obtains an AUC of 0.909. More importantly, the improvement is over contacts at all sequence-position separations. Specifically, a 8.95%, 5.65% and 2.84% increase in precision were observed for the top L\u221510 predictions over the next best for short, medium and long-range contacts, respectively. This confirms the usefulness of ResNets to congregate the short-range relations and 2D-BRLSTM to propagate the long-range dependencies throughout the entire protein contact map \u2018image\u2019.<\/jats:p><\/jats:sec><jats:sec><jats:title>Availability and implementation<\/jats:title><jats:p>SPOT-Contact server url: http:\/\/sparks-lab.org\/jack\/server\/SPOT-Contact\/.<\/jats:p><\/jats:sec><jats:sec><jats:title>Supplementary information<\/jats:title><jats:p>Supplementary data are available at Bioinformatics online.<\/jats:p><\/jats:sec>","DOI":"10.1093\/bioinformatics\/bty481","type":"journal-article","created":{"date-parts":[[2018,6,13]],"date-time":"2018-06-13T11:13:21Z","timestamp":1528888401000},"page":"4039-4045","source":"Crossref","is-referenced-by-count":178,"title":["Accurate prediction of protein contact maps by coupling residual two-dimensional bidirectional long short-term memory with convolutional neural networks"],"prefix":"10.1093","volume":"34","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-6956-6748","authenticated-orcid":false,"given":"Jack","family":"Hanson","sequence":"first","affiliation":[{"name":"Signal Processing Laboratory, Griffith University, Brisbane, Australia"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Kuldip","family":"Paliwal","sequence":"additional","affiliation":[{"name":"Signal Processing Laboratory, Griffith University, Brisbane, Australia"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Thomas","family":"Litfin","sequence":"additional","affiliation":[{"name":"Institute for Glycomics and School of Information and Communication Technology, Griffith University, Southport, Australia"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6782-2813","authenticated-orcid":false,"given":"Yuedong","family":"Yang","sequence":"additional","affiliation":[{"name":"Institute for Glycomics and School of Information and Communication Technology, Griffith University, Southport, Australia"},{"name":"School of Data and Computer Science, Sun-Yat Sen University, Guangzhou, Guangdong, China"}],"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, Australia"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"286","published-online":{"date-parts":[[2018,6,19]]},"reference":[{"key":"2023012712300627600_bty481-B1","first-page":"Abs\/1603.04467","article-title":"Tensorflow: large-scale machine learning on heterogeneous distributed systems","author":"Abadi","year":"2016","journal-title":"CoRR"},{"key":"2023012712300627600_bty481-B2","first-page":"7","article-title":"DNCON2: improved protein contact prediction using two-level deep convolutional neural networks","volume":"1","author":"Adhikari","year":"2017","journal-title":"Bioinformatics"},{"key":"2023012712300627600_bty481-B3","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":"2023012712300627600_bty481-B4","first-page":"602","article-title":"The principled design of large-scale recursive neural network architectures\u2013dag-rnns and the protein structure prediction problem","volume":"575","author":"Baldi","year":"2003","journal-title":"J. Mach. Learn. Res"},{"key":"2023012712300627600_bty481-B5","doi-asserted-by":"crossref","first-page":"361","DOI":"10.1110\/ps.8.2.361","article-title":"Pair potentials for protein folding: choice of reference states and sensitivity of predicted native states to variations in the interaction schemes","volume":"8","author":"Betancourt","year":"1999","journal-title":"Protein Sci"},{"key":"2023012712300627600_bty481-B6","doi-asserted-by":"crossref","first-page":"i75","DOI":"10.1093\/bioinformatics\/bti1004","article-title":"Three-stage prediction of protein \u03b2-sheets by neural networks, alignments and graph algorithms","volume":"21","author":"Cheng","year":"2005","journal-title":"Bioinformatics"},{"key":"2023012712300627600_bty481-B7","doi-asserted-by":"crossref","first-page":"113.","DOI":"10.1186\/1471-2105-8-113","article-title":"Improved residue contact prediction using support vector machines and a large feature set","volume":"8","author":"Cheng","year":"2007","journal-title":"BMC Bioinformatics"},{"key":"2023012712300627600_bty481-B8","article-title":"Fast and accurate deep network learning by exponential linear units (elus","author":"Clevert","year":"2015","journal-title":"arXiv Preprint arXiv: 1511.07289"},{"key":"2023012712300627600_bty481-B9","doi-asserted-by":"crossref","first-page":"2449","DOI":"10.1093\/bioinformatics\/bts475","article-title":"Deep architectures for protein contact map prediction","volume":"28","author":"Di Lena","year":"2012","journal-title":"Bioinformatics"},{"key":"2023012712300627600_bty481-B10","doi-asserted-by":"crossref","first-page":"012707.","DOI":"10.1103\/PhysRevE.87.012707","article-title":"Improved contact prediction in proteins: using pseudolikelihoods to infer potts models","volume":"87","author":"Ekeberg","year":"2013","journal-title":"Phys. Rev. E"},{"key":"2023012712300627600_bty481-B11","doi-asserted-by":"crossref","first-page":"196","DOI":"10.1002\/prot.22554","article-title":"Assessment of domain boundary predictions and the prediction of intramolecular contacts in CASP8","volume":"77","author":"Ezkurdia","year":"2009","journal-title":"Prot. Struct. Func. Bioinform"},{"key":"2023012712300627600_bty481-B12","doi-asserted-by":"crossref","first-page":"861","DOI":"10.1016\/j.patrec.2005.10.010","article-title":"An introduction to ROC analysis","volume":"27","author":"Fawcett","year":"2006","journal-title":"Patt. Recogn. Lett"},{"key":"2023012712300627600_bty481-B13","doi-asserted-by":"crossref","first-page":"309","DOI":"10.1002\/prot.340180402","article-title":"Correlated mutations and residue contacts in proteins","volume":"18","author":"G\u00f6bel","year":"1994","journal-title":"Prot. Struct. Funct. Bioinform"},{"key":"2023012712300627600_bty481-B14","article-title":"Multi-dimensional recurrent neural networks","author":"Graves","year":"2007","journal-title":"CoRR"},{"key":"2023012712300627600_bty481-B15","doi-asserted-by":"crossref","first-page":"29","DOI":"10.1148\/radiology.143.1.7063747","article-title":"The meaning and use of the area under a receiver operating characteristic (ROC) curve","volume":"143","author":"Hanley","year":"1982","journal-title":"Radiology"},{"key":"2023012712300627600_bty481-B16","doi-asserted-by":"crossref","first-page":"993","DOI":"10.1109\/34.58871","article-title":"Neural network ensembles","volume":"12","author":"Hansen","year":"1990","journal-title":"IEEE Trans. Patt. Anal. Mach. Intel"},{"key":"2023012712300627600_bty481-B17","doi-asserted-by":"crossref","first-page":"685","DOI":"10.1093\/bioinformatics\/btw678","article-title":"Improving protein disorder prediction by deep bidirectional long short-term memory recurrent neural networks","volume":"33","author":"Hanson","year":"2017","journal-title":"Bioinformatics"},{"key":"2023012712300627600_bty481-B18","doi-asserted-by":"crossref","first-page":"2296","DOI":"10.1093\/bioinformatics\/btx164","article-title":"NeBcon: protein contact map prediction using neural network training coupled with na\u00efve bayes classifiers","volume":"33","author":"He","year":"2017","journal-title":"Bioinformatics"},{"key":"2023012712300627600_bty481-B19","first-page":"770","volume-title":"Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition","author":"He","year":"2016"},{"key":"2023012712300627600_bty481-B20","first-page":"630","volume-title":"European Conference on Computer Vision.","author":"He","year":"2016"},{"key":"2023012712300627600_bty481-B21","doi-asserted-by":"crossref","first-page":"2842","DOI":"10.1093\/bioinformatics\/btx218","article-title":"Capturing non-local interactions by long short term memory bidirectional recurrent neural networks for improving prediction of protein secondary structure","volume":"33","author":"Heffernan","year":"2017","journal-title":"Bioinformatics"},{"key":"2023012712300627600_bty481-B22","doi-asserted-by":"crossref","first-page":"1527","DOI":"10.1162\/neco.2006.18.7.1527","article-title":"A fast learning algorithm for deep belief nets","volume":"18","author":"Hinton","year":"2006","journal-title":"Neural Comput"},{"key":"2023012712300627600_bty481-B23","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":"2023012712300627600_bty481-B24","doi-asserted-by":"crossref","first-page":"513","DOI":"10.1093\/protein\/8.6.513","article-title":"Identification and analysis of domains in proteins","volume":"8","author":"Islam","year":"1995","journal-title":"Prot. Eng"},{"key":"2023012712300627600_bty481-B25","doi-asserted-by":"crossref","first-page":"184","DOI":"10.1093\/bioinformatics\/btr638","article-title":"PSICOV: precise structural contact prediction using sparse inverse covariance estimation on large multiple sequence alignments","volume":"28","author":"Jones","year":"2012","journal-title":"Bioinformatics"},{"key":"2023012712300627600_bty481-B26","doi-asserted-by":"crossref","first-page":"999","DOI":"10.1093\/bioinformatics\/btu791","article-title":"MetaPSICOV: combining coevolution methods for accurate prediction of contacts and long range hydrogen bonding in proteins","volume":"31","author":"Jones","year":"2015","journal-title":"Bioinformatics"},{"key":"2023012712300627600_bty481-B27","doi-asserted-by":"crossref","first-page":"85.","DOI":"10.1186\/1471-2105-15-85","article-title":"FreeContact: fast and free software for protein contact prediction from residue co-evolution","volume":"15","author":"Kaj\u00e1n","year":"2014","journal-title":"BMC Bioinformatics"},{"key":"2023012712300627600_bty481-B28","doi-asserted-by":"crossref","first-page":"15674","DOI":"10.1073\/pnas.1314045110","article-title":"Assessing the utility of coevolution-based residue\u2013residue contact predictions in a sequence-and structure-rich era","volume":"110","author":"Kamisetty","year":"2013","journal-title":"Proc. Natl. Acad. Sci"},{"key":"2023012712300627600_bty481-B29","article-title":"Residual LSTM: design of a deep recurrent architecture for distant speech recognition","author":"Kim","year":"2017","journal-title":"CoRR"},{"key":"2023012712300627600_bty481-B30","article-title":"Adam: a method for stochastic optimization","author":"Kingma","year":"2014","journal-title":"CoRR"},{"key":"2023012712300627600_bty481-B31","doi-asserted-by":"crossref","first-page":"541","DOI":"10.1162\/neco.1989.1.4.541","article-title":"Backpropagation applied to handwritten zip code recognition","volume":"1","author":"LeCun","year":"1989","journal-title":"Neural Comput"},{"key":"2023012712300627600_bty481-B32","article-title":"Layer normalization","author":"Lei Ba","year":"2016","journal-title":"ArXiv e-Prints, Abs\/1607.06450"},{"key":"2023012712300627600_bty481-B33","doi-asserted-by":"crossref","first-page":"3774","DOI":"10.1093\/bioinformatics\/btw553","article-title":"Kscons: a bayesian approach for protein residue contact prediction using the knob-socket model of protein tertiary structure","volume":"32","author":"Li","year":"2016","journal-title":"Bioinformatics"},{"key":"2023012712300627600_bty481-B34","doi-asserted-by":"crossref","first-page":"360","DOI":"10.1007\/s008940100038","article-title":"Generation and evaluation of dimension-reduced amino acid parameter representations by artificial neural networks","volume":"7","author":"Meiler","year":"2001","journal-title":"Mol. Model. Annu"},{"key":"2023012712300627600_bty481-B35","first-page":"851","article-title":"Deep learning in bioinformatics","volume":"18","author":"Min","year":"2017","journal-title":"Brief. Bioinformatics"},{"key":"2023012712300627600_bty481-B36","doi-asserted-by":"crossref","first-page":"D170","DOI":"10.1093\/nar\/gkw1081","article-title":"Uniclust databases of clustered and deeply annotated protein sequences and alignments","volume":"45","author":"Mirdita","year":"2016","journal-title":"Nucleic Acids Res"},{"key":"2023012712300627600_bty481-B37","doi-asserted-by":"crossref","first-page":"534","DOI":"10.1021\/ma00145a039","article-title":"Estimation of effective interresidue contact energies from protein crystal structures: quasi-chemical approximation","volume":"18","author":"Miyazawa","year":"1985","journal-title":"Macromolecules"},{"key":"2023012712300627600_bty481-B38","doi-asserted-by":"crossref","first-page":"E1293","DOI":"10.1073\/pnas.1111471108","article-title":"Direct-coupling analysis of residue coevolution captures native contacts across many protein families","volume":"108","author":"Morcos","year":"2011","journal-title":"Proc. Natl. Acad. Sci"},{"key":"2023012712300627600_bty481-B39","doi-asserted-by":"crossref","first-page":"1311","DOI":"10.1016\/j.patcog.2004.01.013","article-title":"GPU implementation of neural networks","volume":"37","author":"Oh","year":"2004","journal-title":"Patt. Recogn"},{"key":"2023012712300627600_bty481-B40","doi-asserted-by":"crossref","first-page":"294","DOI":"10.1126\/science.aah4043","article-title":"Protein structure determination using metagenome sequence data","volume":"355","author":"Ovchinnikov","year":"2017","journal-title":"Science"},{"key":"2023012712300627600_bty481-B41","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":"2023012712300627600_bty481-B42","article-title":"Learning internal representations by error propagation","author":"Rumelhart","year":"1985","journal-title":"Tech. Rep. DTIC Document"},{"key":"2023012712300627600_bty481-B43","doi-asserted-by":"crossref","first-page":"51","DOI":"10.1002\/prot.25407","article-title":"Assessment of contact predictions in casp12: co-evolution and deep learning coming of age","volume":"86","author":"Schaarschmidt","year":"2018","journal-title":"Prot. Struct. Funct. Bioinform"},{"key":"2023012712300627600_bty481-B44","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":"2023012712300627600_bty481-B45","doi-asserted-by":"crossref","first-page":"3128","DOI":"10.1093\/bioinformatics\/btu500","article-title":"CCMpredfast and precise prediction of protein residue\u2013residue contacts from correlated mutations","volume":"30","author":"Seemayer","year":"2014","journal-title":"Bioinformatics"},{"key":"2023012712300627600_bty481-B46","doi-asserted-by":"crossref","first-page":"59","DOI":"10.1145\/2983402.2983406","volume-title":"Proceedings of the Third International Symposium on Computer Vision and the Internet VisionNet\u201916","author":"Shah","year":"2016"},{"key":"2023012712300627600_bty481-B47","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":"2023012712300627600_bty481-B48","doi-asserted-by":"crossref","first-page":"W515","DOI":"10.1093\/nar\/gkp305","article-title":"NNcon: improved protein contact map prediction using 2d-recursive neural networks","volume":"37","author":"Tegge","year":"2009","journal-title":"Nucleic Acids Res"},{"key":"2023012712300627600_bty481-B49","volume-title":"Statistical Learning Theory","author":"Vapnik","year":"1998"},{"key":"2023012712300627600_bty481-B50","article-title":"ReNet: a recurrent neural network based alternative to convolutional networks","author":"Visin","year":"2015","journal-title":"CoRR, Abs\/1505.00393"},{"key":"2023012712300627600_bty481-B51","doi-asserted-by":"crossref","first-page":"e1005324","DOI":"10.1371\/journal.pcbi.1005324","article-title":"Accurate de novo prediction of protein contact map by ultra-deep learning model","volume":"13","author":"Wang","year":"2017","journal-title":"PLoS Comput. Biol"},{"key":"2023012712300627600_bty481-B52","doi-asserted-by":"crossref","first-page":"67","DOI":"10.1002\/prot.25377","article-title":"Analysis of deep learning methods for blind protein contact prediction in casp12","volume":"86","author":"Wang","year":"2018","journal-title":"Prot. Struct. Funct. Bioinform"},{"key":"2023012712300627600_bty481-B53","doi-asserted-by":"crossref","first-page":"i266","DOI":"10.1093\/bioinformatics\/btt211","article-title":"Predicting protein contact map using evolutionary and physical constraints by integer programming","volume":"29","author":"Wang","year":"2013","journal-title":"Bioinformatics"},{"key":"2023012712300627600_bty481-B54","doi-asserted-by":"crossref","first-page":"924","DOI":"10.1093\/bioinformatics\/btn069","article-title":"A comprehensive assessment of sequence-based and template-based methods for protein contact prediction","volume":"24","author":"Wu","year":"2008","journal-title":"Bioinformatics"},{"key":"2023012712300627600_bty481-B55","first-page":"219","article-title":"A large-scale comparative assessment of methods for residue\u2013residue contact prediction","volume":"19","author":"Wuyun","year":"2016","journal-title":"Brief. Bioinform"},{"key":"2023012712300627600_bty481-B56","doi-asserted-by":"crossref","first-page":"2675","DOI":"10.1093\/bioinformatics\/btx296","article-title":"A deep learning framework for improving long-range residueresidue contact prediction using a hierarchical strategy","volume":"33","author":"Xiong","year":"2017","journal-title":"Bioinformatics"},{"key":"2023012712300627600_bty481-B57","doi-asserted-by":"crossref","first-page":"2435","DOI":"10.1093\/bioinformatics\/btw181","article-title":"R2C: improving ab initio residue contact map prediction using dynamic fusion strategy and Gaussian noise filter","volume":"32","author":"Yang","year":"2016","journal-title":"Bioinformatics"}],"container-title":["Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/34\/23\/4039\/48920817\/bioinformatics_34_23_4039.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/34\/23\/4039\/48920817\/bioinformatics_34_23_4039.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,9,3]],"date-time":"2023-09-03T04:54:39Z","timestamp":1693716879000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article\/34\/23\/4039\/5040307"}},"subtitle":[],"editor":[{"given":"Alfonso","family":"Valencia","sequence":"additional","affiliation":[],"role":[{"role":"editor","vocabulary":"crossref"}]}],"short-title":[],"issued":{"date-parts":[[2018,6,19]]},"references-count":57,"journal-issue":{"issue":"23","published-print":{"date-parts":[[2018,12,1]]}},"URL":"https:\/\/doi.org\/10.1093\/bioinformatics\/bty481","relation":{},"ISSN":["1367-4803","1367-4811"],"issn-type":[{"value":"1367-4803","type":"print"},{"value":"1367-4811","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2018,12,1]]},"published":{"date-parts":[[2018,6,19]]}}}