{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,1]],"date-time":"2026-04-01T06:45:08Z","timestamp":1775025908649,"version":"3.50.1"},"reference-count":42,"publisher":"Oxford University Press (OUP)","issue":"14","license":[{"start":{"date-parts":[[2019,7,8]],"date-time":"2019-07-08T00:00:00Z","timestamp":1562544000000},"content-version":"vor","delay-in-days":7,"URL":"http:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2019,7,15]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:sec>\n                  <jats:title>Motivation<\/jats:title>\n                  <jats:p>Deep learning architectures have recently demonstrated their power in predicting DNA- and RNA-binding specificity. Existing methods fall into three classes: Some are based on convolutional neural networks (CNNs), others use recurrent neural networks (RNNs) and others rely on hybrid architectures combining CNNs and RNNs. However, based on existing studies the relative merit of the various architectures remains unclear.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Results<\/jats:title>\n                  <jats:p>In this study we present a systematic exploration of deep learning architectures for predicting DNA- and RNA-binding specificity. For this purpose, we present deepRAM, an end-to-end deep learning tool that provides an implementation of a wide selection of architectures; its fully automatic model selection procedure allows us to perform a fair and unbiased comparison of deep learning architectures. We find that deeper more complex architectures provide a clear advantage with sufficient training data, and that hybrid CNN\/RNN architectures outperform other methods in terms of accuracy. Our work provides guidelines that can assist the practitioner in choosing an appropriate network architecture, and provides insight on the difference between the models learned by convolutional and recurrent networks. In particular, we find that although recurrent networks improve model accuracy, this comes at the expense of a loss in the interpretability of the features learned by the model.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Availability and implementation<\/jats:title>\n                  <jats:p>The source code for deepRAM is available at https:\/\/github.com\/MedChaabane\/deepRAM.<\/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\/btz339","type":"journal-article","created":{"date-parts":[[2019,5,14]],"date-time":"2019-05-14T11:42:11Z","timestamp":1557834131000},"page":"i269-i277","source":"Crossref","is-referenced-by-count":150,"title":["Comprehensive evaluation of deep learning architectures for prediction of DNA\/RNA sequence binding specificities"],"prefix":"10.1093","volume":"35","author":[{"given":"Ameni","family":"Trabelsi","sequence":"first","affiliation":[{"name":"Department of Computer Science, Colorado State University, Fort Collins, CO, USA"}]},{"given":"Mohamed","family":"Chaabane","sequence":"additional","affiliation":[{"name":"Department of Computer Science, Colorado State University, Fort Collins, CO, USA"}]},{"given":"Asa","family":"Ben-Hur","sequence":"additional","affiliation":[{"name":"Department of Computer Science, Colorado State University, Fort Collins, CO, USA"}]}],"member":"286","published-online":{"date-parts":[[2019,7,5]]},"reference":[{"key":"2023062712353654100_btz339-B1","doi-asserted-by":"crossref","first-page":"831","DOI":"10.1038\/nbt.3300","article-title":"Predicting the sequence specificities of DNA- and RNA-binding proteins by deep learning","volume":"33","author":"Alipanahi","year":"2015","journal-title":"Nat. Biotechnol"},{"key":"2023062712353654100_btz339-B2","doi-asserted-by":"crossref","first-page":"878.","DOI":"10.15252\/msb.20156651","article-title":"Deep learning for computational biology","volume":"12","author":"Angermueller","year":"2016","journal-title":"Mol. Syst. Biol"},{"key":"2023062712353654100_btz339-B3","doi-asserted-by":"crossref","first-page":"e0141287.","DOI":"10.1371\/journal.pone.0141287","article-title":"Continuous distributed representation of biological sequences for deep proteomics and genomics","volume":"10","author":"Asgari","year":"2015","journal-title":"PLoS One"},{"key":"2023062712353654100_btz339-B4","doi-asserted-by":"crossref","first-page":"157","DOI":"10.1109\/72.279181","article-title":"Learning long-term dependencies with gradient descent is difficult","volume":"5","author":"Bengio","year":"1994","journal-title":"IEEE Trans. Neural Netw"},{"key":"2023062712353654100_btz339-B5","doi-asserted-by":"crossref","first-page":"D160","DOI":"10.1093\/nar\/gku1180","article-title":"DoRiNA 2.0\u2014upgrading the DoRiNA database of RNA interactions in post-transcriptional regulation","volume":"43","author":"Blin","year":"2015","journal-title":"Nucleic Acids Res"},{"key":"2023062712353654100_btz339-B6","doi-asserted-by":"crossref","first-page":"609","DOI":"10.1038\/nmeth.1985","article-title":"Systematic evaluation of factors influencing ChIP-seq fidelity","volume":"9","author":"Chen","year":"2012","journal-title":"Nat. Methods"},{"key":"2023062712353654100_btz339-B7","author":"Cho","year":"2014"},{"key":"2023062712353654100_btz339-B8","author":"Chung","year":"2014"},{"key":"2023062712353654100_btz339-B9","doi-asserted-by":"crossref","first-page":"1188","DOI":"10.1101\/gr.849004","article-title":"WebLogo: a sequence logo generator","volume":"14","author":"Crooks","year":"2004","journal-title":"Genome Res"},{"key":"2023062712353654100_btz339-B10","doi-asserted-by":"crossref","first-page":"57","DOI":"10.1038\/nature11247","article-title":"An integrated Encyclopedia of DNA elements in the human genome","volume":"489","year":"2012","journal-title":"Nature"},{"key":"2023062712353654100_btz339-B11","doi-asserted-by":"crossref","first-page":"106","DOI":"10.1093\/bib\/bbv031","article-title":"Revealing protein-lncRNA interaction","volume":"17","author":"Ferr\u00e9","year":"2016","journal-title":"Brief. Bioinform"},{"key":"2023062712353654100_btz339-B12","doi-asserted-by":"crossref","first-page":"829","DOI":"10.1038\/nrg3813","article-title":"A census of human RNA-binding proteins","volume":"15","author":"Gerstberger","year":"2014","journal-title":"Nat. Rev. Genet"},{"key":"2023062712353654100_btz339-B13","author":"Gupta","year":"2017"},{"key":"2023062712353654100_btz339-B14","doi-asserted-by":"crossref","first-page":"R24","DOI":"10.1186\/gb-2007-8-2-r24","article-title":"Quantifying similarity between motifs","volume":"8","author":"Gupta","year":"2007","journal-title":"Genome Biol"},{"key":"2023062712353654100_btz339-B15","doi-asserted-by":"crossref","first-page":"178","DOI":"10.1109\/BIBM.2016.7822515","volume-title":"2016 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), Shenzhen, China","author":"Hassanzadeh","year":"2016"},{"key":"2023062712353654100_btz339-B16","doi-asserted-by":"crossref","first-page":"82","DOI":"10.1109\/MSP.2012.2205597","article-title":"Deep neural networks for acoustic modeling in speech recognition: the shared views of four research groups","volume":"29","author":"Hinton","year":"2012","journal-title":"IEEE Signal Process. Mag"},{"key":"2023062712353654100_btz339-B17","doi-asserted-by":"crossref","first-page":"261","DOI":"10.1126\/science.aaa8685","article-title":"Advances in natural language processing","volume":"349","author":"Hirschberg","year":"2015","journal-title":"Science"},{"key":"2023062712353654100_btz339-B18","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":"2023062712353654100_btz339-B19","doi-asserted-by":"crossref","first-page":"e1000832.","DOI":"10.1371\/journal.pcbi.1000832","article-title":"RNAcontext: a new method for learning the sequence and structure binding preferences of RNA-binding proteins","volume":"6","author":"Kazan","year":"2010","journal-title":"PLoS Comput. Biol"},{"key":"2023062712353654100_btz339-B20","doi-asserted-by":"crossref","first-page":"739","DOI":"10.1101\/gr.227819.117","article-title":"Sequential regulatory activity prediction across chromosomes with convolutional neural networks","volume":"28","author":"Kelley","year":"2018","journal-title":"Genome Res"},{"key":"2023062712353654100_btz339-B21","first-page":"1097","article-title":"ImageNet classification with deep convolutional neural networks","author":"Krizhevsky","year":"2012","journal-title":"Advances in neural information processing systems"},{"key":"2023062712353654100_btz339-B22","doi-asserted-by":"crossref","first-page":"2278","DOI":"10.1109\/5.726791","article-title":"Gradient-based learning applied to document recognition","volume":"86","author":"LeCun","year":"1998","journal-title":"Proc. IEEE"},{"key":"2023062712353654100_btz339-B23","doi-asserted-by":"crossref","first-page":"436","DOI":"10.1038\/nature14539","article-title":"Deep learning","volume":"521","author":"LeCun","year":"2015","journal-title":"Nature"},{"key":"2023062712353654100_btz339-B24","author":"Lipton","year":"2018"},{"key":"2023062712353654100_btz339-B25","article-title":"Rectifier nonlinearities improve neural network acoustic models","author":"Maas","year":"2013","journal-title":"ICML Workshop on Deep Learning for Audio, Speech and Language Processing"},{"key":"2023062712353654100_btz339-B26","doi-asserted-by":"crossref","first-page":"D142","DOI":"10.1093\/nar\/gkt997","article-title":"JASPAR 2014: an extensively expanded and updated open-access database of transcription factor binding profiles","volume":"42","author":"Mathelier","year":"2014","journal-title":"Nucleic Acids Res"},{"key":"2023062712353654100_btz339-B27","author":"Melis","year":"2018"},{"key":"2023062712353654100_btz339-B28","first-page":"3111","article-title":"Distributed representations of words and phrases and their compositionality","author":"Mikolov","year":"2013","journal-title":"Adv. Neural Inf. Process. Syst"},{"key":"2023062712353654100_btz339-B29","doi-asserted-by":"crossref","first-page":"i92","DOI":"10.1093\/bioinformatics\/btx234","article-title":"Chromatin accessibility prediction via convolutional long short-term memory networks with k-mer embedding","volume":"33","author":"Min","year":"2017","journal-title":"Bioinformatics"},{"key":"2023062712353654100_btz339-B30","doi-asserted-by":"crossref","first-page":"511.","DOI":"10.1186\/s12864-018-4889-1","article-title":"Prediction of RNA-protein sequence and structure binding preferences using deep convolutional and recurrent neural networks","volume":"19","author":"Pan","year":"2018","journal-title":"BMC Genomics"},{"key":"2023062712353654100_btz339-B31","doi-asserted-by":"crossref","first-page":"e107","DOI":"10.1093\/nar\/gkw226","article-title":"DanQ: a hybrid convolutional and recurrent deep neural network for quantifying the function of DNA sequences","volume":"44","author":"Quang","year":"2016","journal-title":"Nucleic Acids Res"},{"key":"2023062712353654100_btz339-B32","doi-asserted-by":"crossref","first-page":"172","DOI":"10.1038\/nature12311","article-title":"A compendium of RNA-binding motifs for decoding gene regulation","volume":"499","author":"Ray","year":"2013","journal-title":"Nature"},{"key":"2023062712353654100_btz339-B33","doi-asserted-by":"crossref","first-page":"233","DOI":"10.1146\/annurev-biochem-060408-091030","article-title":"Origins of specificity in protein-DNA recognition","volume":"79","author":"Rohs","year":"2010","journal-title":"Annu. Rev. Biochem"},{"key":"2023062712353654100_btz339-B34","doi-asserted-by":"crossref","first-page":"15270.","DOI":"10.1038\/s41598-018-33321-1","article-title":"Recurrent neural network for predicting transcription factor binding sites","volume":"8","author":"Shen","year":"2018","journal-title":"Sci. Rep"},{"key":"2023062712353654100_btz339-B35","doi-asserted-by":"crossref","first-page":"2099","DOI":"10.1093\/nar\/gkt1112","article-title":"Protein-DNA binding: complexities and multi-protein codes","volume":"42","author":"Siggers","year":"2014","journal-title":"Nucleic Acids Res"},{"key":"2023062712353654100_btz339-B36","doi-asserted-by":"crossref","first-page":"16","DOI":"10.1093\/bioinformatics\/16.1.16","article-title":"DNA binding sites: representation and discovery","volume":"16","author":"Stormo","year":"2000","journal-title":"Bioinformatics"},{"key":"2023062712353654100_btz339-B37","doi-asserted-by":"crossref","first-page":"1527","DOI":"10.1093\/bioinformatics\/btw003","article-title":"Orthogonal matrix factorization enables integrative analysis of multiple RNA binding proteins","volume":"32","author":"Stra\u017ear","year":"2016","journal-title":"Bioinformatics"},{"key":"2023062712353654100_btz339-B38","author":"Strubell","year":"2017"},{"key":"2023062712353654100_btz339-B39","first-page":"3104","article-title":"Sequence to sequence learning with neural networks","author":"Sutskever","year":"2014","journal-title":"Adv. Neural Inf. Process. Syst"},{"key":"2023062712353654100_btz339-B40","author":"Yu","year":"2015"},{"key":"2023062712353654100_btz339-B41","doi-asserted-by":"crossref","first-page":"i121","DOI":"10.1093\/bioinformatics\/btw255","article-title":"Convolutional neural network architectures for predicting DNA-protein binding","volume":"32","author":"Zeng","year":"2016","journal-title":"Bioinformatics"},{"key":"2023062712353654100_btz339-B42","doi-asserted-by":"crossref","first-page":"931","DOI":"10.1038\/nmeth.3547","article-title":"Predicting effects of noncoding variants with deep learning-based sequence model","volume":"12","author":"Zhou","year":"2015","journal-title":"Nat. Methods"}],"container-title":["Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/35\/14\/i269\/50721273\/bioinformatics_35_14_i269.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/35\/14\/i269\/50721273\/bioinformatics_35_14_i269.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,6,27]],"date-time":"2023-06-27T12:36:26Z","timestamp":1687869386000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article\/35\/14\/i269\/5529112"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2019,7]]},"references-count":42,"journal-issue":{"issue":"14","published-print":{"date-parts":[[2019,7,15]]}},"URL":"https:\/\/doi.org\/10.1093\/bioinformatics\/btz339","relation":{},"ISSN":["1367-4803","1367-4811"],"issn-type":[{"value":"1367-4803","type":"print"},{"value":"1367-4811","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2019,7]]},"published":{"date-parts":[[2019,7]]}}}