{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,26]],"date-time":"2026-02-26T20:34:44Z","timestamp":1772138084862,"version":"3.50.1"},"reference-count":41,"publisher":"Oxford University Press (OUP)","issue":"8","license":[{"start":{"date-parts":[[2017,11,16]],"date-time":"2017-11-16T00:00:00Z","timestamp":1510790400000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100001659","name":"Deutsche Forschungsgemeinschaft","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100001659","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100007065","name":"Nvidia","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100007065","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2018,4,15]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Motivation<\/jats:title>\n                    <jats:p>Regulatory sequences are not solely defined by their nucleic acid sequence but also by their relative distances to genomic landmarks such as transcription start site, exon boundaries or polyadenylation site. Deep learning has become the approach of choice for modeling regulatory sequences because of its strength to learn complex sequence features. However, modeling relative distances to genomic landmarks in deep neural networks has not been addressed.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>Here we developed spline transformation, a neural network module based on splines to flexibly and robustly model distances. Modeling distances to various genomic landmarks with spline transformations significantly increased state-of-the-art prediction accuracy of in vivo RNA-binding protein binding sites for 120 out of 123 proteins. We also developed a deep neural network for human splice branchpoint based on spline transformations that outperformed the current best, already distance-based, machine learning model. Compared to piecewise linear transformation, as obtained by composition of rectified linear units, spline transformation yields higher prediction accuracy as well as faster and more robust training. As spline transformation can be applied to further quantities beyond distances, such as methylation or conservation, we foresee it as a versatile component in the genomics deep learning toolbox.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Availability and implementation<\/jats:title>\n                    <jats:p>Spline transformation is implemented as a Keras layer in the CONCISE python package: https:\/\/github.com\/gagneurlab\/concise. Analysis code is available at https:\/\/github.com\/gagneurlab\/Manuscript_Avsec_Bioinformatics_2017.<\/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\/btx727","type":"journal-article","created":{"date-parts":[[2017,11,15]],"date-time":"2017-11-15T15:28:58Z","timestamp":1510759738000},"page":"1261-1269","source":"Crossref","is-referenced-by-count":31,"title":["Modeling positional effects of regulatory sequences with spline transformations increases prediction accuracy of deep neural networks"],"prefix":"10.1093","volume":"34","author":[{"given":"\u017diga","family":"Avsec","sequence":"first","affiliation":[{"name":"Department of Informatics, Technical University of Munich, Garching, Germany"},{"name":"Graduate School of Quantitative Biosciences (QBM), Gene Center, Ludwig-Maximilians-Universit\u00e4t M\u00fcnchen, Munich, Germany"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8470-8203","authenticated-orcid":false,"given":"Mohammadamin","family":"Barekatain","sequence":"additional","affiliation":[{"name":"Department of Informatics, Technical University of Munich, Garching, Germany"}]},{"given":"Jun","family":"Cheng","sequence":"additional","affiliation":[{"name":"Department of Informatics, Technical University of Munich, Garching, Germany"},{"name":"Graduate School of Quantitative Biosciences (QBM), Gene Center, Ludwig-Maximilians-Universit\u00e4t M\u00fcnchen, Munich, Germany"}]},{"given":"Julien","family":"Gagneur","sequence":"additional","affiliation":[{"name":"Department of Informatics, Technical University of Munich, Garching, Germany"}]}],"member":"286","published-online":{"date-parts":[[2017,11,16]]},"reference":[{"key":"2023012713010737100_btx727-B1","author":"Abadi","year":"2016"},{"key":"2023012713010737100_btx727-B2","author":"Alexandari","year":"2017"},{"key":"2023012713010737100_btx727-B3","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":"2023012713010737100_btx727-B4","doi-asserted-by":"crossref","first-page":"67.","DOI":"10.1186\/s13059-017-1189-z","article-title":"DeepCpG: accurate prediction of single-cell DNA methylation states using deep learning","volume":"18","author":"Angermueller","year":"2017","journal-title":"Genome Biol"},{"key":"2023012713010737100_btx727-B5","author":"Bastien","year":"2012"},{"key":"2023012713010737100_btx727-B6","first-page":"115","author":"Bergstra","year":"2013"},{"key":"2023012713010737100_btx727-B7","doi-asserted-by":"crossref","first-page":"1169","DOI":"10.1101\/gr.166819.113","article-title":"LaSSO, a strategy for genome-wide mapping of intronic lariats and branch points using RNA-seq","volume":"24","author":"Bitton","year":"2014","journal-title":"Genome Res"},{"key":"2023012713010737100_btx727-B8","doi-asserted-by":"crossref","first-page":"1534","DOI":"10.1126\/science.3952495","article-title":"Heterogeneous nuclear ribonucleoproteins: role in RNA splicing","volume":"231","author":"Choi","year":"1986","journal-title":"Science"},{"key":"2023012713010737100_btx727-B9","author":"Chollet","year":"2015"},{"key":"2023012713010737100_btx727-B10","author":"Collobert","year":"2002"},{"key":"2023012713010737100_btx727-B11","doi-asserted-by":"crossref","first-page":"e1001016","DOI":"10.1371\/journal.pcbi.1001016","article-title":"Genome-wide association between branch point properties and alternative splicing","volume":"6","author":"Corvelo","year":"2010","journal-title":"PLoS Comput. Biol"},{"key":"2023012713010737100_btx727-B12","author":"De Boor","year":"1978"},{"key":"2023012713010737100_btx727-B13","doi-asserted-by":"crossref","first-page":"89","DOI":"10.1214\/ss\/1038425655","article-title":"Flexible smoothing with B-splines and penalties","volume":"11","author":"Eilers","year":"1996","journal-title":"Stat. Sci"},{"key":"2023012713010737100_btx727-B14","doi-asserted-by":"crossref","first-page":"636","DOI":"10.1126\/science.1105136","article-title":"The ENCODE (ENCyclopedia Of DNA Elements) Project","volume":"306","author":"ENCODE Project Consortium","year":"2004","journal-title":"Science"},{"key":"2023012713010737100_btx727-B15","doi-asserted-by":"crossref","first-page":"1","DOI":"10.18637\/jss.v033.i01","article-title":"Regularization paths for generalized linear models via coordinate descent","volume":"33","author":"Friedman","year":"2010","journal-title":"J. Stat. Softw"},{"key":"2023012713010737100_btx727-B16","doi-asserted-by":"crossref","first-page":"2257","DOI":"10.1093\/nar\/gkn073","article-title":"Human branch point consensus sequence is yUnAy","volume":"36","author":"Gao","year":"2008","journal-title":"Nucleic Acids Res"},{"key":"2023012713010737100_btx727-B17","doi-asserted-by":"crossref","first-page":"1760","DOI":"10.1101\/gr.135350.111","article-title":"GENCODE: the reference human genome annotation for The ENCODE Project","volume":"22","author":"Harrow","year":"2012","journal-title":"Genome Res"},{"key":"2023012713010737100_btx727-B18","volume-title":"Generalized Additive Models","author":"Hastie","year":"1990"},{"key":"2023012713010737100_btx727-B19","first-page":"448","volume-title":"Proceedings of the 32nd International Conference on Machine Learning","author":"Ioffe","year":"2015"},{"key":"2023012713010737100_btx727-B20","doi-asserted-by":"crossref","first-page":"675","DOI":"10.1145\/2647868.2654889","volume-title":"Proceedings of the 22nd ACM International Conference on Multimedia","author":"Jia","year":"2014"},{"key":"2023012713010737100_btx727-B21","doi-asserted-by":"crossref","first-page":"990","DOI":"10.1101\/gr.200535.115","article-title":"Basset: learning the regulatory code of the accessible genome with deep convolutional neural networks","volume":"26","author":"Kelley","year":"2016","journal-title":"Genome Res"},{"key":"2023012713010737100_btx727-B22","author":"Kingma","year":"2014"},{"key":"2023012713010737100_btx727-B23","first-page":"1","volume-title":"J. Stat. Softw","author":"Kuhn","year":"2008"},{"key":"2023012713010737100_btx727-B24","doi-asserted-by":"crossref","first-page":"i121","DOI":"10.1093\/bioinformatics\/btu277","article-title":"Deep learning of the tissue-regulated splicing code","volume":"30","author":"Leung","year":"2014","journal-title":"Bioinformatics"},{"key":"2023012713010737100_btx727-B25","doi-asserted-by":"crossref","first-page":"290","DOI":"10.1101\/gr.182899.114","article-title":"Genome-wide discovery of human splicing branchpoints","volume":"25","author":"Mercer","year":"2015","journal-title":"Genome Res"},{"key":"2023012713010737100_btx727-B26","first-page":"2924","volume-title":"Advances in neural information processing systems","author":"Mont\u00fafar","year":"2014"},{"key":"2023012713010737100_btx727-B27","first-page":"807","author":"Nair","year":"2010"},{"key":"2023012713010737100_btx727-B28","doi-asserted-by":"crossref","first-page":"136.","DOI":"10.1186\/s12859-017-1561-8","article-title":"RNA-protein binding motifs mining with a new hybrid deep learning based cross-domain knowledge integration approach","volume":"18","author":"Pan","year":"2017","journal-title":"BMC Bioinformatics"},{"key":"2023012713010737100_btx727-B29","doi-asserted-by":"crossref","first-page":"833","DOI":"10.1016\/S0092-8674(85)80064-7","article-title":"Cryptic branch point activation allows accurate in vitro splicing of human \u03b2-globin intron mutants","volume":"41","author":"Ruskin","year":"1985","journal-title":"Cell"},{"key":"2023012713010737100_btx727-B30","doi-asserted-by":"crossref","first-page":"129","DOI":"10.1038\/nrm3952","article-title":"Structural basis of transcription initiation by RNA polymerase II","volume":"16","author":"Sainsbury","year":"2015","journal-title":"Nat. Rev. Mol. Cell Biol"},{"key":"2023012713010737100_btx727-B31","author":"Shrikumar","year":"2017"},{"key":"2023012713010737100_btx727-B32","author":"Signal","year":"2016"},{"key":"2023012713010737100_btx727-B33","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":"2023012713010737100_btx727-B34","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 (Oxford, England)"},{"key":"2023012713010737100_btx727-B35","first-page":"2258","author":"Stricker","year":"2017"},{"key":"2023012713010737100_btx727-B36","first-page":"26","article-title":"Lecture 6.5-rmsprop: divide the gradient by a running average of its recent magnitude","volume":"4","author":"Tieleman","year":"2012","journal-title":"COURSERA: Neural Netw. Mach. Learn"},{"key":"2023012713010737100_btx727-B37","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1038\/nmeth.3810","article-title":"Robust transcriptome-wide discovery of RNA-binding protein binding sites with enhanced CLIP (eCLIP)","volume":"13","author":"Van Nostrand","year":"2016","journal-title":"Nat. Methods"},{"key":"2023012713010737100_btx727-B38","doi-asserted-by":"crossref","first-page":"701","DOI":"10.1016\/j.cell.2009.02.009","article-title":"The spliceosome: design principles of a dynamic RNP machine","volume":"136","author":"Wahl","year":"2009","journal-title":"Cell"},{"key":"2023012713010737100_btx727-B39","doi-asserted-by":"crossref","DOI":"10.1201\/9781420010404","volume-title":"Generalized Additive Models: An Introduction with R","author":"Wood","year":"2006"},{"key":"2023012713010737100_btx727-B40","doi-asserted-by":"crossref","first-page":"1254806","DOI":"10.1126\/science.1254806","article-title":"The human splicing code reveals new insights into the genetic determinants of disease","volume":"347","author":"Xiong","year":"2015","journal-title":"Science"},{"key":"2023012713010737100_btx727-B41","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\/34\/8\/1261\/48915392\/bioinformatics_34_8_1261.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/34\/8\/1261\/48915392\/bioinformatics_34_8_1261.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,1,27]],"date-time":"2023-01-27T08:51:25Z","timestamp":1674809485000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article\/34\/8\/1261\/4636216"}},"subtitle":[],"editor":[{"given":"Inanc","family":"Birol","sequence":"additional","affiliation":[]}],"short-title":[],"issued":{"date-parts":[[2017,11,16]]},"references-count":41,"journal-issue":{"issue":"8","published-print":{"date-parts":[[2018,4,15]]}},"URL":"https:\/\/doi.org\/10.1093\/bioinformatics\/btx727","relation":{"has-preprint":[{"id-type":"doi","id":"10.1101\/165183","asserted-by":"object"}]},"ISSN":["1367-4803","1367-4811"],"issn-type":[{"value":"1367-4803","type":"print"},{"value":"1367-4811","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2018,4,15]]},"published":{"date-parts":[[2017,11,16]]}}}