000151616 001__ 151616
000151616 005__ 20240321221037.0
000151616 0247_ $$2doi$$a10.1089/cmb.2019.0320
000151616 0247_ $$2pmid$$apmid:31855058
000151616 0247_ $$2pmc$$apmc:PMC7047095
000151616 0247_ $$2ISSN$$a1066-5277
000151616 0247_ $$2ISSN$$a1557-8666
000151616 037__ $$aDZNE-2020-01198
000151616 041__ $$aEnglish
000151616 082__ $$a570
000151616 1001_ $$0P:(DE-HGF)0$$aFiosina, Jelena$$b0
000151616 245__ $$aExplainable Deep Learning for Augmentation of Small RNA Expression Profiles.342
000151616 260__ $$aLarchmont, NY$$bLiebert$$c2019
000151616 3367_ $$2DRIVER$$aarticle
000151616 3367_ $$2DataCite$$aOutput Types/Journal article
000151616 3367_ $$0PUB:(DE-HGF)16$$2PUB:(DE-HGF)$$aJournal Article$$bjournal$$mjournal$$s1601292570_15840
000151616 3367_ $$2BibTeX$$aARTICLE
000151616 3367_ $$2ORCID$$aJOURNAL_ARTICLE
000151616 3367_ $$00$$2EndNote$$aJournal Article
000151616 520__ $$aThe lack of well-structured metadata annotations complicates the reusability and interpretation of the growing amount of publicly available RNA expression data. The machine learning-based prediction of metadata (data augmentation) can considerably improve the quality of expression data annotation. In this study, we systematically benchmark deep learning (DL) and random forest (RF)-based metadata augmentation of tissue, age, and sex using small RNA (sRNA) expression profiles. We use 4243 annotated sRNA-Seq samples from the sRNA expression atlas database to train and test the augmentation performance. In general, the DL machine learner outperforms the RF method in almost all tested cases. The average cross-validated prediction accuracy of the DL algorithm for tissues is 96.5%, for sex is 77%, and for age is 77.2%. The average tissue prediction accuracy for a completely new data set is 83.1% (DL) and 80.8% (RF). To understand which sRNAs influence DL predictions, we employ backpropagation-based feature importance scores using the DeepLIFT method, which enable us to obtain information on biological relevance of sRNAs.
000151616 536__ $$0G:(DE-HGF)POF3-342$$a342 - Disease Mechanisms and Model Systems (POF3-342)$$cPOF3-342$$fPOF III$$x0
000151616 588__ $$aDataset connected to CrossRef, PubMed,
000151616 7001_ $$0P:(DE-2719)2811935$$aFiosins, Maksims$$b1$$eCorresponding author$$udzne
000151616 7001_ $$0P:(DE-2719)2810547$$aBonn, Stefan$$b2$$eLast author$$udzne
000151616 773__ $$0PERI:(DE-600)2030900-4$$a10.1089/cmb.2019.0320$$gVol. 27, no. 2, p. 234 - 247$$n2$$p234 - 247$$tJournal of computational biology$$v27$$x1557-8666$$y2019
000151616 909CO $$ooai:pub.dzne.de:151616$$pVDB
000151616 9101_ $$0I:(DE-588)1065079516$$6P:(DE-2719)2811935$$aDeutsches Zentrum für Neurodegenerative Erkrankungen$$b1$$kDZNE
000151616 9101_ $$0I:(DE-588)1065079516$$6P:(DE-2719)2810547$$aDeutsches Zentrum für Neurodegenerative Erkrankungen$$b2$$kDZNE
000151616 9131_ $$0G:(DE-HGF)POF3-342$$1G:(DE-HGF)POF3-340$$2G:(DE-HGF)POF3-300$$aDE-HGF$$bForschungsbereich Gesundheit$$lErkrankungen des Nervensystems$$vDisease Mechanisms and Model Systems$$x0
000151616 9141_ $$y2019
000151616 915__ $$0StatID:(DE-HGF)0160$$2StatID$$aDBCoverage$$bEssential Science Indicators$$d2019-12-20
000151616 915__ $$0StatID:(DE-HGF)1190$$2StatID$$aDBCoverage$$bBiological Abstracts$$d2019-12-20
000151616 915__ $$0StatID:(DE-HGF)0420$$2StatID$$aNationallizenz$$d2023-03-30$$wger
000151616 915__ $$0StatID:(DE-HGF)0100$$2StatID$$aJCR$$bJ COMPUT BIOL : 2021$$d2023-03-30
000151616 915__ $$0StatID:(DE-HGF)0200$$2StatID$$aDBCoverage$$bSCOPUS$$d2023-03-30
000151616 915__ $$0StatID:(DE-HGF)0300$$2StatID$$aDBCoverage$$bMedline$$d2023-03-30
000151616 915__ $$0StatID:(DE-HGF)0600$$2StatID$$aDBCoverage$$bEbsco Academic Search$$d2023-03-30
000151616 915__ $$0StatID:(DE-HGF)0030$$2StatID$$aPeer Review$$bASC$$d2023-03-30
000151616 915__ $$0StatID:(DE-HGF)0199$$2StatID$$aDBCoverage$$bClarivate Analytics Master Journal List$$d2023-03-30
000151616 915__ $$0StatID:(DE-HGF)0150$$2StatID$$aDBCoverage$$bWeb of Science Core Collection$$d2023-03-30
000151616 915__ $$0StatID:(DE-HGF)1050$$2StatID$$aDBCoverage$$bBIOSIS Previews$$d2023-03-30
000151616 915__ $$0StatID:(DE-HGF)1030$$2StatID$$aDBCoverage$$bCurrent Contents - Life Sciences$$d2023-03-30
000151616 915__ $$0StatID:(DE-HGF)9900$$2StatID$$aIF < 5$$d2023-03-30
000151616 9201_ $$0I:(DE-2719)1410003$$kAG Bonn 1$$lComputational analysis of biological networks$$x0
000151616 980__ $$ajournal
000151616 980__ $$aVDB
000151616 980__ $$aI:(DE-2719)1410003
000151616 980__ $$aUNRESTRICTED