{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,3]],"date-time":"2026-04-03T15:02:24Z","timestamp":1775228544892,"version":"3.50.1"},"reference-count":57,"publisher":"Oxford University Press (OUP)","issue":"10","license":[{"start":{"date-parts":[[2016,10,12]],"date-time":"2016-10-12T00:00:00Z","timestamp":1476230400000},"content-version":"vor","delay-in-days":268,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2016,5,15]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:p>Motivation: RNA binding proteins (RBPs) play important roles in post-transcriptional control of gene expression, including splicing, transport, polyadenylation and RNA stability. To model protein\u2013RNA interactions by considering all available sources of information, it is necessary to integrate the rapidly growing RBP experimental data with the latest genome annotation, gene function, RNA sequence and structure. Such integration is possible by matrix factorization, where current approaches have an undesired tendency to identify only a small number of the strongest patterns with overlapping features. Because protein\u2013RNA interactions are orchestrated by multiple factors, methods that identify discriminative patterns of varying strengths are needed.<\/jats:p>\n               <jats:p>Results: We have developed an integrative orthogonality-regularized nonnegative matrix factorization (iONMF) to integrate multiple data sources and discover non-overlapping, class-specific RNA binding patterns of varying strengths. The orthogonality constraint halves the effective size of the factor model and outperforms other NMF models in predicting RBP interaction sites on RNA. We have integrated the largest data compendium to date, which includes 31 CLIP experiments on 19 RBPs involved in splicing (such as hnRNPs, U2AF2, ELAVL1, TDP-43 and FUS) and processing of 3\u2019UTR (Ago, IGF2BP). We show that the integration of multiple data sources improves the predictive accuracy of retrieval of RNA binding sites. In our study the key predictive factors of protein\u2013RNA interactions were the position of RNA structure and sequence motifs, RBP co-binding and gene region type. We report on a number of protein-specific patterns, many of which are consistent with experimentally determined properties of RBPs.<\/jats:p>\n               <jats:p>Availability and implementation: The iONMF implementation and example datasets are available at https:\/\/github.com\/mstrazar\/ionmf.<\/jats:p>\n               <jats:p>Contact: tomaz.curk@fri.uni-lj.si<\/jats:p>\n               <jats:p>Supplementary information: \u00a0Supplementary data are available at Bioinformatics online.<\/jats:p>","DOI":"10.1093\/bioinformatics\/btw003","type":"journal-article","created":{"date-parts":[[2016,2,15]],"date-time":"2016-02-15T01:09:07Z","timestamp":1455498547000},"page":"1527-1535","source":"Crossref","is-referenced-by-count":118,"title":["Orthogonal matrix factorization enables integrative analysis of multiple RNA binding proteins"],"prefix":"10.1093","volume":"32","author":[{"given":"Martin","family":"Stra\u017ear","sequence":"first","affiliation":[{"name":"1 University of Ljubljana, Faculty of Computer and Information Science, Ljubljana, SI 1000, Slovenia"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Marinka","family":"\u017ditnik","sequence":"additional","affiliation":[{"name":"1 University of Ljubljana, Faculty of Computer and Information Science, Ljubljana, SI 1000, Slovenia"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Bla\u017e","family":"Zupan","sequence":"additional","affiliation":[{"name":"1 University of Ljubljana, Faculty of Computer and Information Science, Ljubljana, SI 1000, Slovenia"},{"name":"2 Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, TX 77030, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jernej","family":"Ule","sequence":"additional","affiliation":[{"name":"3 Department of Molecular Neuroscience, UCL Institute of Neurology, Queen Square, London WC1N 3BG, UK"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Toma\u017e","family":"Curk","sequence":"additional","affiliation":[{"name":"1 University of Ljubljana, Faculty of Computer and Information Science, Ljubljana, SI 1000, Slovenia"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"286","published-online":{"date-parts":[[2016,1,18]]},"reference":[{"key":"2023020112243872800_btw003-B1","doi-asserted-by":"crossref","first-page":"543","DOI":"10.1016\/j.celrep.2012.03.013","article-title":"Differential GC content between exons and introns establishes distinct strategies of splice-site recognition","volume":"1","author":"Amit","year":"2012","journal-title":"Cell Reports"},{"key":"2023020112243872800_btw003-B2","doi-asserted-by":"crossref","first-page":"D180","DOI":"10.1093\/nar\/gkr1007","article-title":"doRiNA: a database of RNA interactions in post-transcriptional regulation","volume":"40","author":"Anders","year":"2012","journal-title":"Nucleic Acids Res"},{"key":"2023020112243872800_btw003-B3","doi-asserted-by":"crossref","first-page":"25","DOI":"10.1038\/75556","article-title":"Gene Ontology: tool for the unification of biology","volume":"25","author":"Ashburner","year":"2000","journal-title":"Nat. Genet"},{"key":"2023020112243872800_btw003-B4","doi-asserted-by":"crossref","first-page":"1247","DOI":"10.1101\/gr.073155.107","article-title":"A systematic analysis of intronic sequences downstream of 5\u2019 splice sites reveals a widespread role for U-rich motifs and TIA1\/TIAL1 proteins in alternative splicing regulation","volume":"18","author":"Aznarez","year":"2008","journal-title":"Genome Res"},{"key":"2023020112243872800_btw003-B5","doi-asserted-by":"crossref","first-page":"294","DOI":"10.1016\/j.neuron.2013.10.062","article-title":"Transcriptome-wide discovery of microRNA binding sites in human brain","volume":"81","author":"Boudreau","year":"2014","journal-title":"Neuron"},{"key":"2023020112243872800_btw003-B6","doi-asserted-by":"crossref","first-page":"4164","DOI":"10.1073\/pnas.0308531101","article-title":"Metagenes and molecular pattern discovery using matrix factorization","volume":"101","author":"Brunet","year":"2004","journal-title":"Proc. Natl. Acad. Sci. U.S.A"},{"key":"2023020112243872800_btw003-B7","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1186\/1471-2105-7-78","article-title":"Biclustering of gene expression data by non-smooth non-negative matrix factorization","volume":"7","author":"Carmona-Saez","year":"2006","journal-title":"BMC Bioinf"},{"key":"2023020112243872800_btw003-B8","doi-asserted-by":"crossref","first-page":"1393","DOI":"10.1016\/j.cell.2012.04.031","article-title":"Insights into RNA biology from an atlas of mammalian mRNA-binding proteins","volume":"149","author":"Castello","year":"2012","journal-title":"Cell"},{"key":"2023020112243872800_btw003-B9","doi-asserted-by":"crossref","first-page":"2370","DOI":"10.1101\/gad.250993.114","article-title":"CPSF30 and Wdr33 directly bind to AAUAAA in mammalian mRNA 3 processing","volume":"28","author":"Chan","year":"2014","journal-title":"Genes Dev"},{"key":"2023020112243872800_btw003-B10","volume-title":"Regression Analysis by Example","author":"Chatterjee","year":"2015"},{"key":"2023020112243872800_btw003-B11","doi-asserted-by":"crossref","first-page":"161","DOI":"10.1002\/wcms.1119","article-title":"Predictions of protein\u2013RNA interactions","volume":"3","author":"Cirillo","year":"2013","journal-title":"Comp. Mol. Sci"},{"key":"2023020112243872800_btw003-B12","doi-asserted-by":"crossref","first-page":"243","DOI":"10.1146\/annurev-neuro-062912-114322","article-title":"RNA protein interaction in neurons","volume":"36","author":"Darnell","year":"2013","journal-title":"Annu. Rev. Neurosci"},{"key":"2023020112243872800_btw003-B13","first-page":"1","article-title":"Statistical comparisons of classifiers over multiple data sets","volume":"7","author":"Dem\u0161ar","year":"2006","journal-title":"J. Mach. Learn. Res"},{"key":"2023020112243872800_btw003-B14","first-page":"1090","article-title":"Using RNAFOLD to predict the activity of small catalytic RNAs","volume":"15","author":"Denman","year":"1993","journal-title":"BioTechniques"},{"key":"2023020112243872800_btw003-B15","doi-asserted-by":"crossref","first-page":"107","DOI":"10.1002\/wrna.1147","article-title":"Computational biology of RNA interactions","volume":"4","author":"Dieterich","year":"2013","journal-title":"Wiley Interdisc. Rev. RNA"},{"key":"2023020112243872800_btw003-B16","author":"Ding","year":"2006"},{"key":"2023020112243872800_btw003-B17","doi-asserted-by":"crossref","first-page":"45","DOI":"10.1109\/TPAMI.2008.277","article-title":"Convex and semi-nonnegative matrix factorizations","volume":"32","author":"Ding","year":"2010","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell"},{"key":"2023020112243872800_btw003-B18","doi-asserted-by":"crossref","first-page":"3970","DOI":"10.1093\/bioinformatics\/bti653","article-title":"Improving molecular cancer class discovery through sparse non-negative matrix factorization","volume":"21","author":"Gao","year":"2005","journal-title":"Bioinformatics"},{"key":"2023020112243872800_btw003-B19","doi-asserted-by":"crossref","first-page":"1087","DOI":"10.1109\/TNNLS.2012.2197827","article-title":"Online nonnegative matrix factorization with robust stochastic approximation","volume":"23","author":"Guan","year":"2012","journal-title":"IEEE Trans. Neural Netw. Learn. Syst"},{"key":"2023020112243872800_btw003-B20","doi-asserted-by":"crossref","first-page":"129","DOI":"10.1016\/j.cell.2010.03.009","article-title":"Transcriptome-wide identification of RNA-binding protein and microRNA target sites by PAR-CLIP","volume":"141","author":"Hafner","year":"2010","journal-title":"Cell"},{"key":"2023020112243872800_btw003-B21","doi-asserted-by":"crossref","first-page":"1428","DOI":"10.1038\/nsmb.2163","article-title":"RNA targets of wild-type and mutant FET family proteins","volume":"18","author":"Hoell","year":"2011","journal-title":"Nat. Struct. Mol. Biol"},{"key":"2023020112243872800_btw003-B22","first-page":"1457","article-title":"Non-negative matrix factorization with sparseness constraints","volume":"5","author":"Hoyer","year":"2004","journal-title":"J. Mach. Learn. Res"},{"key":"2023020112243872800_btw003-B23","doi-asserted-by":"crossref","first-page":"38","DOI":"10.1093\/nar\/30.1.38","article-title":"The Ensembl genome database project","volume":"30","author":"Hubbard","year":"2002","journal-title":"Nucleic Acids Res"},{"key":"2023020112243872800_btw003-B24","doi-asserted-by":"crossref","first-page":"255","DOI":"10.1016\/j.celrep.2013.06.029","article-title":"NSun2-mediated cytosine-5 methylation of vault noncoding RNA determines its processing into regulatory small RNAs","volume":"4","author":"Hussain","year":"2013","journal-title":"Cell Rep"},{"key":"2023020112243872800_btw003-B25","doi-asserted-by":"crossref","first-page":"2684","DOI":"10.1093\/bioinformatics\/btn526","article-title":"Position-dependent motif characterization using non-negative matrix factorization","volume":"24","author":"Hutchins","year":"2008","journal-title":"Bioinformatics"},{"key":"2023020112243872800_btw003-B26","doi-asserted-by":"crossref","first-page":"113","DOI":"10.1038\/nrg3853","article-title":"Competition between target sites of regulators shapes post-transcriptional gene regulation","volume":"16","author":"Jens","year":"2015","journal-title":"Nat. Rev. Genet"},{"key":"2023020112243872800_btw003-B27","author":"Joshi","year":"2010"},{"key":"2023020112243872800_btw003-B28","doi-asserted-by":"crossref","first-page":"1","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":"2023020112243872800_btw003-B29","doi-asserted-by":"crossref","first-page":"1495","DOI":"10.1093\/bioinformatics\/btm134","article-title":"Sparse non-negative matrix factorizations via alternating non-negativity-constrained least squares for microarray data analysis","volume":"23","author":"Kim","year":"2007","journal-title":"Bioinformatics"},{"key":"2023020112243872800_btw003-B30","doi-asserted-by":"crossref","first-page":"559","DOI":"10.1038\/nmeth.1608","article-title":"A quantitative analysis of CLIP methods for identifying binding sites of RNA-binding proteins","volume":"8","author":"Kishore","year":"2011","journal-title":"Nat. Methods"},{"key":"2023020112243872800_btw003-B31","doi-asserted-by":"crossref","first-page":"1601","DOI":"10.1093\/bioinformatics\/btu074","article-title":"The cleverSuite approach for protein characterization: predictions of structural properties, solubility, chaperone requirements and RNA-binding abilities","volume":"30","author":"Klus","year":"2014","journal-title":"Bioinformatics"},{"key":"2023020112243872800_btw003-B32","doi-asserted-by":"crossref","first-page":"909","DOI":"10.1038\/nsmb.1838","article-title":"iCLIP reveals the function of hnRNP particles in splicing at individual nucleotide resolution","volume":"17","author":"K\u00f6nig","year":"2010","journal-title":"Nat. Struct. Mol. Biol"},{"key":"2023020112243872800_btw003-B33","first-page":"548","article-title":"Algorithms for non-negative matrix factorization","author":"Lee","year":"2001","journal-title":"Adv. NIPS"},{"key":"2023020112243872800_btw003-B34","doi-asserted-by":"crossref","first-page":"1096","DOI":"10.1261\/rna.2017210","article-title":"Predicting in vivo binding sites of RNA-binding proteins using mRNA secondary structure","author":"Li","year":"2010","journal-title":"RNA"},{"key":"2023020112243872800_btw003-B35","doi-asserted-by":"crossref","first-page":"2756","DOI":"10.1162\/neco.2007.19.10.2756","article-title":"Projected gradient methods for nonnegative matrix factorization","volume":"19","author":"Lin","year":"2007","journal-title":"Neural Comput"},{"key":"2023020112243872800_btw003-B36","doi-asserted-by":"crossref","first-page":"123","DOI":"10.1186\/1471-2105-15-123","article-title":"Protein-specific prediction of mRNA binding using RNA sequences, binding motifs and pred. secondary structures","volume":"15","author":"Livi","year":"2014","journal-title":"BMC Bioinf"},{"key":"2023020112243872800_btw003-B37","doi-asserted-by":"crossref","first-page":"R17","DOI":"10.1186\/gb-2014-15-1-r17","article-title":"GraphProt: modeling binding preferences of RNA-binding proteins","volume":"15","author":"Maticzka","year":"2014","journal-title":"Genome Biol"},{"key":"2023020112243872800_btw003-B38","doi-asserted-by":"crossref","first-page":"1884","DOI":"10.1093\/bioinformatics\/btp331","article-title":"Predictor correlation impacts machine learning algorithms: implications for genomic studies","volume":"25","author":"Nicodemus","year":"2009","journal-title":"Bioinformatics"},{"key":"2023020112243872800_btw003-B39","doi-asserted-by":"crossref","first-page":"261","DOI":"10.1016\/j.jsb.2011.10.001","article-title":"Computational methods for prediction of protein\u2013RNA interactions","volume":"179","author":"Puton","year":"2012","journal-title":"J. Struct. Biol"},{"key":"2023020112243872800_btw003-B40","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":"2023020112243872800_btw003-B41","doi-asserted-by":"crossref","first-page":"146","DOI":"10.4161\/rna.27991","article-title":"Crosslinking-immunoprecipitation (iCLIP) analysis reveals global regulatory roles of hnRNP L","volume":"11","author":"Rossbach","year":"2014","journal-title":"RNA Biol"},{"key":"2023020112243872800_btw003-B42","doi-asserted-by":"crossref","first-page":"381","DOI":"10.1101\/gr.082503.108","article-title":"Splicing factor SFRS1 recognizes a functionally diverse landscape of RNA transcripts","volume":"19","author":"Sanford","year":"2009","journal-title":"Genome Res"},{"key":"2023020112243872800_btw003-B43","doi-asserted-by":"crossref","first-page":"1124","DOI":"10.1038\/nsmb.2420","article-title":"CLIP-seq of eIF4AIII reveals transcriptome-wide mapping of the human exon junction complex","volume":"19","author":"Sauli\u00e8re","year":"2012","journal-title":"Nat. Struct. Mol. Biol"},{"key":"2023020112243872800_btw003-B44","doi-asserted-by":"crossref","first-page":"e160","DOI":"10.1093\/nar\/gks697","article-title":"Mixture models and wavelet transforms reveal high confidence RNA\u2013protein interaction sites in MOV10 PAR-CLIP data","volume":"40","author":"Sievers","year":"2012","journal-title":"Nucleic Acids Res"},{"key":"2023020112243872800_btw003-B45","doi-asserted-by":"crossref","first-page":"R67","DOI":"10.1186\/gb-2012-13-8-r67","article-title":"Analysis of CLIP and iCLIP methods for nucleotide-resolution studies of protein\u2013RNA interactions","volume":"13","author":"Sugimoto","year":"2012","journal-title":"Genome Biol"},{"key":"2023020112243872800_btw003-B46","doi-asserted-by":"crossref","first-page":"452","DOI":"10.1038\/nn.2778","article-title":"Characterizing the RNA targets and position-dependent splicing regulation by TDP-43","volume":"14","author":"Tollervey","year":"2011","journal-title":"Nat. Neurosci"},{"key":"2023020112243872800_btw003-B47","doi-asserted-by":"crossref","first-page":"344","DOI":"10.1089\/cmb.2012.0273","article-title":"Predicting protein\u2013protein interactions from multimodal biological data sources via nonnegative matrix tri-factorization","volume":"20","author":"Wang","year":"2013","journal-title":"J. Comput. Biol"},{"key":"2023020112243872800_btw003-B48","doi-asserted-by":"crossref","first-page":"e1000530","DOI":"10.1371\/journal.pbio.1000530","article-title":"iCLIP Predicts the Dual Splicing Effects of TIA-RNA Interactions","volume":"8","author":"Wang","year":"2010","journal-title":"PLoS Biol"},{"key":"2023020112243872800_btw003-B49","first-page":"396","article-title":"Weighted nonnegative matrix co-tri-factorization for collaborative prediction","volume":"5828","author":"Yoo","year":"2009","journal-title":"Adv. Mach. Learn"},{"key":"2023020112243872800_btw003-B50","doi-asserted-by":"crossref","first-page":"453","DOI":"10.1016\/j.cell.2012.12.023","article-title":"Direct competition between hnRNP C and U2AF65 protects the transcriptome from the exonization of Alu elements","volume":"152","author":"Zarnack","year":"2013","journal-title":"Cell"},{"key":"2023020112243872800_btw003-B51","first-page":"870","article-title":"Non-negative matrix factorization with quasi-newton optimization","volume":"87","author":"Zdunek","year":"2006","journal-title":"Artif. Intell. Soft Comput"},{"key":"2023020112243872800_btw003-B52","doi-asserted-by":"crossref","first-page":"439","DOI":"10.1126\/science.1191150","article-title":"Integrative modeling defines the Nova splicing-regulatory network and its combinatorial controls","volume":"329","author":"Zhang","year":"2010","journal-title":"Science"},{"key":"2023020112243872800_btw003-B53","doi-asserted-by":"crossref","first-page":"i401","DOI":"10.1093\/bioinformatics\/btr206","article-title":"A novel computational framework for simultaneous integration of multiple types of genomic data to identify microRNA-gene regulatory modules","volume":"27","author":"Zhang","year":"2011","journal-title":"Bioinformatics"},{"key":"2023020112243872800_btw003-B54","doi-asserted-by":"crossref","first-page":"9379","DOI":"10.1093\/nar\/gks725","article-title":"Discovery of multi-dimensional modules by integrative analysis of cancer genomic data","volume":"40","author":"Zhang","year":"2012","journal-title":"Nucleic Acids Res"},{"key":"2023020112243872800_btw003-B55","doi-asserted-by":"crossref","first-page":"6793","DOI":"10.1093\/nar\/gkt421","article-title":"Prediction of clustered RNA-binding protein motif sites in the mammalian genome","volume":"41","author":"Zhang","year":"2013","journal-title":"Nucleic Acids Res"},{"key":"2023020112243872800_btw003-B56","author":"Zhou","year":"2011"},{"key":"2023020112243872800_btw003-B57","doi-asserted-by":"crossref","first-page":"41","DOI":"10.1109\/TPAMI.2014.2343973","article-title":"Data fusion by matrix factorization","volume":"37","author":"Zitnik","year":"2015","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell"}],"container-title":["Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/32\/10\/1527\/49019169\/bioinformatics_32_10_1527.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/32\/10\/1527\/49019169\/bioinformatics_32_10_1527.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,2,1]],"date-time":"2023-02-01T22:30:22Z","timestamp":1675290622000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article\/32\/10\/1527\/1742711"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2016,1,18]]},"references-count":57,"journal-issue":{"issue":"10","published-print":{"date-parts":[[2016,5,15]]}},"URL":"https:\/\/doi.org\/10.1093\/bioinformatics\/btw003","relation":{},"ISSN":["1367-4811","1367-4803"],"issn-type":[{"value":"1367-4811","type":"electronic"},{"value":"1367-4803","type":"print"}],"subject":[],"published-other":{"date-parts":[[2016,5,15]]},"published":{"date-parts":[[2016,1,18]]}}}