{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,10,30]],"date-time":"2025-10-30T17:31:23Z","timestamp":1761845483999,"version":"3.37.3"},"reference-count":59,"publisher":"Public Library of Science (PLoS)","issue":"2","license":[{"start":{"date-parts":[[2022,2,24]],"date-time":"2022-02-24T00:00:00Z","timestamp":1645660800000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100000038","name":"Natural Sciences and Engineering Research Council of Canada","doi-asserted-by":"publisher","award":["RGPIN-2017-06743"],"award-info":[{"award-number":["RGPIN-2017-06743"]}],"id":[{"id":"10.13039\/501100000038","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["61873089"],"award-info":[{"award-number":["61873089"]}],"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":["62032007"],"award-info":[{"award-number":["62032007"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100004543","name":"Chinese Scholarship Council","doi-asserted-by":"crossref","id":[{"id":"10.13039\/501100004543","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["www.ploscompbiol.org"],"crossmark-restriction":false},"short-container-title":["PLoS Comput Biol"],"abstract":"<jats:p>Precise identification of target sites of RNA-binding proteins (RBP) is important to understand their biochemical and cellular functions. A large amount of experimental data is generated by in vivo and in vitro approaches. The binding preferences determined from these platforms share similar patterns but there are discernable differences between these datasets. Computational methods trained on one dataset do not always work well on another dataset. To address this problem which resembles the classic \u201cdomain shift\u201d in deep learning, we adopted the adversarial domain adaptation (ADDA) technique and developed a framework (RBP-ADDA) that can extract RBP binding preferences from an integration of in vivo and vitro datasets. Compared with conventional methods, ADDA has the advantage of working with two input datasets, as it trains the initial neural network for each dataset individually, projects the two datasets onto a feature space, and uses an adversarial framework to derive an optimal network that achieves an optimal discriminative predictive power. In the first step, for each RBP, we include only the in vitro data to pre-train a source network and a task predictor. Next, for the same RBP, we initiate the target network by using the source network and use adversarial domain adaptation to update the target network using both in vitro and in vivo data. These two steps help leverage the in vitro data to improve the prediction on in vivo data, which is typically challenging with a lower signal-to-noise ratio. Finally, to further take the advantage of the fused source and target data, we fine-tune the task predictor using both data. We showed that RBP-ADDA achieved better performance in modeling in vivo RBP binding data than other existing methods as judged by Pearson correlations. It also improved predictive performance on in vitro datasets. We further applied augmentation operations on RBPs with less in vivo data to expand the input data and showed that it can improve prediction performances. Lastly, we explored the predictive interpretability of RBP-ADDA, where we quantified the contribution of the input features by Integrated Gradients and identified nucleotide positions that are important for RBP recognition.<\/jats:p>","DOI":"10.1371\/journal.pcbi.1009863","type":"journal-article","created":{"date-parts":[[2022,2,24]],"date-time":"2022-02-24T18:32:03Z","timestamp":1645727523000},"page":"e1009863","update-policy":"https:\/\/doi.org\/10.1371\/journal.pcbi.corrections_policy","source":"Crossref","is-referenced-by-count":4,"title":["Inferring RNA-binding protein target preferences using adversarial domain adaptation"],"prefix":"10.1371","volume":"18","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-3740-9144","authenticated-orcid":true,"given":"Ying","family":"Liu","sequence":"first","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4266-6420","authenticated-orcid":true,"given":"Ruihui","family":"Li","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2385-8272","authenticated-orcid":true,"given":"Jiawei","family":"Luo","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7303-6472","authenticated-orcid":true,"given":"Zhaolei","family":"Zhang","sequence":"additional","affiliation":[]}],"member":"340","published-online":{"date-parts":[[2022,2,24]]},"reference":[{"key":"pcbi.1009863.ref001","doi-asserted-by":"crossref","first-page":"829","DOI":"10.1038\/nrg3813","article-title":"A census of human RNA-binding proteins","volume":"15","author":"S Gerstberger","year":"2014","journal-title":"Nature Reviews Genetics"},{"key":"pcbi.1009863.ref002","doi-asserted-by":"crossref","first-page":"777","DOI":"10.1016\/j.cell.2009.02.011","article-title":"RNA and disease","volume":"136","author":"TA Cooper","year":"2009","journal-title":"Cell"},{"key":"pcbi.1009863.ref003","doi-asserted-by":"crossref","first-page":"13","DOI":"10.1002\/wrna.101","article-title":"mRNA export and cancer","volume":"3","author":"N Siddiqui","year":"2012","journal-title":"Wiley Interdiscip Rev RNA"},{"key":"pcbi.1009863.ref004","first-page":"77","article-title":"Protein\u2013RNA interactions: new genomic technologies and perspectives","volume":"13","author":"J K\u00f6nig","year":"2012","journal-title":"Nature Publishing Group"},{"key":"pcbi.1009863.ref005","doi-asserted-by":"crossref","first-page":"266","DOI":"10.1002\/wrna.31","article-title":"HITS-CLIP: panoramic views of protein-RNA regulation in living cells","volume":"1","author":"RB Darnell","year":"2010","journal-title":"Wiley Interdiscip Rev RNA"},{"key":"pcbi.1009863.ref006","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":"M Hafner","year":"2010","journal-title":"Cell"},{"key":"pcbi.1009863.ref007","article-title":"iCLIP\u2014transcriptome-wide mapping of protein-RNA interactions with individual nucleotide resolution","author":"J Konig","year":"2011","journal-title":"J Vis Exp."},{"key":"pcbi.1009863.ref008","doi-asserted-by":"crossref","first-page":"50","DOI":"10.1016\/j.ymeth.2016.12.007","article-title":"CRISPR\/Cas9-mediated integration enables TAG-eCLIP of endogenously tagged RNA binding proteins","volume":"118\u2013119","author":"EL Van Nostrand","year":"2017","journal-title":"Methods"},{"key":"pcbi.1009863.ref009","doi-asserted-by":"crossref","first-page":"508","DOI":"10.1038\/nmeth.3810","article-title":"Robust transcriptome-wide discovery of RNA-binding protein binding sites with enhanced CLIP (eCLIP)","volume":"13","author":"EL Van Nostrand","year":"2016","journal-title":"Nature Methods"},{"key":"pcbi.1009863.ref010","doi-asserted-by":"crossref","first-page":"74","DOI":"10.1093\/bfgp\/elu047","article-title":"High-throughput characterization of protein-RNA interactions","volume":"14","author":"KB Cook","year":"2015","journal-title":"Briefings in Functional Genomics"},{"key":"pcbi.1009863.ref011","doi-asserted-by":"crossref","first-page":"887","DOI":"10.1016\/j.molcel.2014.04.016","article-title":"RNA Bind-n-Seq: quantitative assessment of the sequence and structural binding specificity of RNA binding proteins","volume":"54","author":"N Lambert","year":"2014","journal-title":"Mol Cell"},{"key":"pcbi.1009863.ref012","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":"D Ray","year":"2013","journal-title":"Nature"},{"key":"pcbi.1009863.ref013","doi-asserted-by":"crossref","first-page":"e117","DOI":"10.1093\/nar\/gkl544","article-title":"Using RNA secondary structures to guide sequence motif finding towards single-stranded regions","volume":"34","author":"M Hiller","year":"2006","journal-title":"Nucleic Acids Res"},{"key":"pcbi.1009863.ref014","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":"H Kazan","year":"2010","journal-title":"PLoS Comput Biol"},{"key":"pcbi.1009863.ref015","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":"D Maticzka","year":"2014","journal-title":"Genome Biol"},{"key":"pcbi.1009863.ref016","doi-asserted-by":"crossref","first-page":"3427","DOI":"10.1093\/bioinformatics\/bty364","article-title":"Predicting RNA-protein binding sites and motifs through combining local and global deep convolutional neural networks","volume":"34","author":"X Pan","year":"2018","journal-title":"Bioinformatics"},{"key":"pcbi.1009863.ref017","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":"B Alipanahi","year":"2015","journal-title":"Nature Biotechnology"},{"key":"pcbi.1009863.ref018","doi-asserted-by":"crossref","first-page":"i638","DOI":"10.1093\/bioinformatics\/bty600","article-title":"A deep neural network approach for learning intrinsic protein-RNA binding preferences","volume":"34","author":"I Ben-Bassat","year":"2018","journal-title":"Bioinformatics"},{"key":"pcbi.1009863.ref019","doi-asserted-by":"crossref","first-page":"214","DOI":"10.1101\/gr.247494.118","article-title":"Deep neural networks for interpreting RNA-binding protein target preferences","volume":"30","author":"M Ghanbari","year":"2020","journal-title":"Genome Res"},{"key":"pcbi.1009863.ref020","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":"X Pan","year":"2018","journal-title":"BMC Genomics"},{"key":"pcbi.1009863.ref021","doi-asserted-by":"crossref","first-page":"e83","DOI":"10.1093\/nar\/gkw048","article-title":"Modeling the combined effect of RNA-binding proteins and microRNAs in post-transcriptional regulation","volume":"44","author":"S HafezQorani","year":"2016","journal-title":"Nucleic Acids Res"},{"key":"pcbi.1009863.ref022","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":"S Kishore","year":"2011","journal-title":"Nat Methods"},{"key":"pcbi.1009863.ref023","article-title":"Sequence biases in CLIP experimental data are incorporated in protein RNA-binding models","author":"Y Orenstein","year":"2016","journal-title":"bioRxiv"},{"key":"pcbi.1009863.ref024","article-title":"Mind the Class Weight Bias: Weighted Maximum Mean Discrepancy for Unsupervised Domain Adaptation","author":"H Yan","year":"2017","journal-title":"Computer Vision and Pattern Recognitio (CVPR17)."},{"key":"pcbi.1009863.ref025","article-title":"Conditional Adversarial Domain Adaptation","volume":"31","author":"M Long","year":"2018","journal-title":"Advances in Neural Information Processing Systems"},{"journal-title":"Learning Transferable Features with Deep Adaptation Networks","year":"2015","author":"M Long","key":"pcbi.1009863.ref026"},{"key":"pcbi.1009863.ref027","article-title":") Deep Domain Confusion: Maximizing for Domain Invariance.","author":"E Tzeng","year":"2014","journal-title":"arXiv"},{"key":"pcbi.1009863.ref028","article-title":"Return of Frustratingly Easy Domain Adaptation","author":"B Sun","year":"2016","journal-title":"Thirtieth AAAI Conference on Artificial Intelligence"},{"key":"pcbi.1009863.ref029","article-title":"Deep CORAL: Correlation Alignment for Deep Domain Adaptation","author":"B Sun","year":"2016","journal-title":"arXiv 1607.01719"},{"journal-title":"Deep Reconstruction-Classification Networks for Unsupervised Domain Adaptation","year":"2016","author":"M Ghifary","key":"pcbi.1009863.ref030"},{"journal-title":"Adversarial Discriminative Domain Adaptation","year":"2017","author":"E Tzeng","key":"pcbi.1009863.ref031"},{"key":"pcbi.1009863.ref032","doi-asserted-by":"crossref","first-page":"i260","DOI":"10.1093\/bioinformatics\/btz364","article-title":"Adversarial domain adaptation for cross data source macromolecule in situ structural classification in cellular electron cryo-tomograms","volume":"35","author":"R Lin","year":"2019","journal-title":"Bioinformatics"},{"key":"pcbi.1009863.ref033","doi-asserted-by":"crossref","first-page":"325","DOI":"10.1109\/JBHI.2020.3032060","article-title":"Measuring Domain Shift for Deep Learning in Histopathology","volume":"25","author":"K Stacke","year":"2021","journal-title":"IEEE Journal of Biomedical and Health Informatics"},{"key":"pcbi.1009863.ref034","doi-asserted-by":"crossref","first-page":"3458","DOI":"10.1038\/s41467-020-17281-7","article-title":"Searching large-scale scRNA-seq databases via unbiased cell embedding with Cell BLAST","volume":"11","author":"ZJ Cao","year":"2020","journal-title":"Nat Commun"},{"key":"pcbi.1009863.ref035","doi-asserted-by":"crossref","first-page":"5131","DOI":"10.1038\/s41467-020-18918-3","article-title":"Deep transfer learning for reducing health care disparities arising from biomedical data inequality","volume":"11","author":"Y Gao","year":"2020","journal-title":"Nat Commun"},{"key":"pcbi.1009863.ref036","doi-asserted-by":"crossref","first-page":"2973","DOI":"10.1093\/bioinformatics\/bty190","article-title":"Generalizing biomedical relation classification with neural adversarial domain adaptation","volume":"34","author":"A Rios","year":"2018","journal-title":"Bioinformatics"},{"key":"pcbi.1009863.ref037","first-page":"3319","article-title":"Axiomatic Attribution for Deep Networks.","author":"M Sundararajan","year":"2017","journal-title":"Proceedings of the 34th International Conference on Machine Learning, PMLR"},{"key":"pcbi.1009863.ref038","doi-asserted-by":"crossref","first-page":"711","DOI":"10.1038\/s41586-020-2077-3","article-title":"A large-scale binding and functional map of human RNA-binding proteins","volume":"583","author":"EL Van Nostrand","year":"2020","journal-title":"Nature"},{"key":"pcbi.1009863.ref039","doi-asserted-by":"crossref","first-page":"26","DOI":"10.1186\/1748-7188-6-26","article-title":"ViennaRNA Package 2.0.","volume":"6","author":"R Lorenz","year":"2011","journal-title":"Algorithms Mol Biol"},{"key":"pcbi.1009863.ref040","article-title":"Rectified Linear Units Improve Restricted Boltzmann Machines.","author":"V Nair","year":"2010","journal-title":"27th International Conference on Machine Learning (ICML-10). Haifa, Israel"},{"key":"pcbi.1009863.ref041","doi-asserted-by":"crossref","first-page":"151","DOI":"10.1007\/s10994-009-5152-4","article-title":"A theory of learning from different domains","volume":"79","author":"S Ben-David","year":"2010","journal-title":"Machine Learning"},{"key":"pcbi.1009863.ref042","article-title":"EDA: Easy Data Augmentation Techniques for Boosting Performance on Text Classification Tasks","author":"J Wei","year":"2019","journal-title":"arXiv 1901.11196"},{"key":"pcbi.1009863.ref043","doi-asserted-by":"crossref","first-page":"11381","DOI":"10.1007\/s00500-019-04602-2","article-title":"Data augmentation using MG-GAN for improved cancer classification on gene expression data","volume":"24","author":"P Chaudhari","year":"2019","journal-title":"Soft Computing"},{"key":"pcbi.1009863.ref044","first-page":"1803","article-title":"How to Explain Individual Classification Decisions","volume":"11","author":"D Baehrens","year":"2010","journal-title":"Journal of Machine Learning Research"},{"key":"pcbi.1009863.ref045","article-title":"Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps","author":"K Simonyan","year":"2014","journal-title":"arXiv 1312.6034"},{"key":"pcbi.1009863.ref046","doi-asserted-by":"crossref","first-page":"3645","DOI":"10.1093\/bioinformatics\/btx469","article-title":"ggseqlogo: a versatile R package for drawing sequence logos","volume":"33","author":"O Wagih","year":"2017","journal-title":"Bioinformatics"},{"key":"pcbi.1009863.ref047","doi-asserted-by":"crossref","first-page":"3001","DOI":"10.1016\/j.jmb.2015.05.020","article-title":"The Signature of the Five-Stranded vRRM Fold Defined by Functional, Structural and Computational Analysis of the hnRNP L Protein","volume":"427","author":"M Blatter","year":"2015","journal-title":"J Mol Biol"},{"key":"pcbi.1009863.ref048","doi-asserted-by":"crossref","first-page":"1443","DOI":"10.1038\/nsmb.2698","article-title":"Molecular basis of UG-rich RNA recognition by the human splicing factor TDP-43","volume":"20","author":"PJ Lukavsky","year":"2013","journal-title":"Nat Struct Mol Biol"},{"key":"pcbi.1009863.ref049","doi-asserted-by":"crossref","first-page":"708","DOI":"10.1128\/MCB.23.2.708-720.2003","article-title":"Heterogeneous nuclear ribonucleoprotein C modulates translation of c-myc mRNA in a cell cycle phase-dependent manner","volume":"23","author":"JH Kim","year":"2003","journal-title":"Mol Cell Biol"},{"key":"pcbi.1009863.ref050","article-title":"Enhancer Identification using Transfer and Adversarial Deep Learning of DNA Sequences","author":"D Cohn","year":"2018","journal-title":"bioRxiv"},{"key":"pcbi.1009863.ref051","doi-asserted-by":"crossref","first-page":"492","DOI":"10.1016\/j.ajhg.2019.01.018","article-title":"Discovery of Allele-Specific Protein-RNA Interactions in Human Transcriptomes","volume":"104","author":"E Bahrami-Samani","year":"2019","journal-title":"Am J Hum Genet"},{"key":"pcbi.1009863.ref052","doi-asserted-by":"crossref","first-page":"515094","DOI":"10.3389\/fgene.2020.515094","article-title":"A Survey of Regulatory Interactions Among RNA Binding Proteins and MicroRNAs in Cancer.","volume":"11","author":"Y Liu","year":"2020","journal-title":"Front Genet"},{"key":"pcbi.1009863.ref053","doi-asserted-by":"crossref","first-page":"327","DOI":"10.1038\/nrm.2017.130","article-title":"A brave new world of RNA-binding proteins","volume":"19","author":"MW Hentze","year":"2018","journal-title":"Nat Rev Mol Cell Biol"},{"key":"pcbi.1009863.ref054","doi-asserted-by":"crossref","first-page":"322","DOI":"10.1038\/s41594-019-0200-7","article-title":"RNA structure maps across mammalian cellular compartments","volume":"26","author":"L Sun","year":"2019","journal-title":"Nat Struct Mol Biol"},{"key":"pcbi.1009863.ref055","doi-asserted-by":"crossref","first-page":"4958","DOI":"10.1093\/nar\/gkz250","article-title":"Discovering sequence and structure landscapes in RNA interaction motifs","volume":"47","author":"M Adinolfi","year":"2019","journal-title":"Nucleic Acids Res"},{"key":"pcbi.1009863.ref056","article-title":"RNANetMotif: identifying sequence-structure RNA network motifs in RNA-protein binding sites","author":"H Ma","year":"2021","journal-title":"bioRxiv"},{"key":"pcbi.1009863.ref057","doi-asserted-by":"crossref","first-page":"1242","DOI":"10.1038\/nbt.3343","article-title":"Affinity regression predicts the recognition code of nucleic acid-binding proteins","volume":"33","author":"R Pelossof","year":"2015","journal-title":"Nat Biotechnol"},{"key":"pcbi.1009863.ref058","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":"J Zhou","year":"2015","journal-title":"Nat Methods"},{"key":"pcbi.1009863.ref059","doi-asserted-by":"crossref","first-page":"166","DOI":"10.1038\/s41588-020-00761-3","article-title":"Genome-wide landscape of RNA-binding protein target site dysregulation reveals a major impact on psychiatric disorder risk","volume":"53","author":"CY Park","year":"2021","journal-title":"Nat Genet"}],"container-title":["PLOS Computational Biology"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pcbi.1009863","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,2,24]],"date-time":"2022-02-24T18:33:05Z","timestamp":1645727585000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pcbi.1009863"}},"subtitle":[],"editor":[{"given":"Yasser","family":"Roudi","sequence":"first","affiliation":[]}],"short-title":[],"issued":{"date-parts":[[2022,2,24]]},"references-count":59,"journal-issue":{"issue":"2","published-online":{"date-parts":[[2022,2,24]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pcbi.1009863","relation":{},"ISSN":["1553-7358"],"issn-type":[{"type":"electronic","value":"1553-7358"}],"subject":[],"published":{"date-parts":[[2022,2,24]]}}}