{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,13]],"date-time":"2026-03-13T18:52:16Z","timestamp":1773427936521,"version":"3.50.1"},"reference-count":26,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2021,11,24]],"date-time":"2021-11-24T00:00:00Z","timestamp":1637712000000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2021,11,24]],"date-time":"2021-11-24T00:00:00Z","timestamp":1637712000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"name":"Brainwaves Northern Ireland Charity"},{"name":"Department for the Economy Northern Ireland"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["BMC Bioinformatics"],"published-print":{"date-parts":[[2021,12]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:sec>\n                <jats:title>Background<\/jats:title>\n                <jats:p>Liver cancer (Hepatocellular carcinoma; HCC) prevalence is increasing and with poor clinical outcome expected it means greater understanding of HCC aetiology is urgently required. This study explored a deep learning solution to detect biologically important features that distinguish prognostic subgroups. A novel architecture of an Artificial Neural Network (ANN) trained with a customised objective function (L<jats:sub>RSC<\/jats:sub>) was developed. The ANN should discover new data representations, to detect patient subgroups that are biologically homogenous (clustering loss) and similar in survival (survival loss) while removing noise from the data (reconstruction loss). The model was applied to TCGA-HCC multi-omics data and benchmarked against baseline models that only use a reconstruction objective function (BCE, MSE) for learning. With the baseline models, the new features are then filtered based on survival information and used for clustering patients. Different variants of the customised objective function, incorporating only reconstruction and clustering losses (L<jats:sub>RC<\/jats:sub>); and reconstruction and survival losses (L<jats:sub>RS<\/jats:sub>) were also evaluated. Robust features consistently detected were compared between models and validated in TCGA and LIRI-JP HCC cohorts.<\/jats:p>\n              <\/jats:sec><jats:sec>\n                <jats:title>Results<\/jats:title>\n                <jats:p>The combined loss (L<jats:sub>RSC<\/jats:sub>) discovered highly significant prognostic subgroups (<jats:italic>P<\/jats:italic>-value\u2009=\u20091.55E\u221277) with more accurate sample assignment (Silhouette scores: 0.59\u20130.7) compared to baseline models (0.18\u20130.3). All L<jats:sub>RSC<\/jats:sub> bottleneck features (N\u2009=\u2009100) were significant for survival, compared to only 11\u201321 for baseline models. Prognostic subgroups were not explained by disease grade or risk factors. Instead L<jats:sub>RSC<\/jats:sub> identified robust features including 377 mRNAs, many of which were novel (61.27%) compared to those identified by the other losses. Some 75 mRNAs were prognostic in TCGA, while 29 were prognostic in LIRI-JP also. L<jats:sub>RSC<\/jats:sub> also identified 15 robust miRNAs including two novel (hsa-let-7g; hsa-mir-550a-1) and 328 methylation features with 71% being prognostic. Gene-enrichment and Functional Annotation Analysis identified seven pathways differentiating prognostic clusters.<\/jats:p>\n              <\/jats:sec><jats:sec>\n                <jats:title>Conclusions<\/jats:title>\n                <jats:p>Combining cluster and survival metrics with the reconstruction objective function facilitated superior prognostic subgroup identification. The hybrid model identified more homogeneous clusters that consequently were more biologically meaningful. The novel and prognostic robust features extracted provide additional information to improve our understanding of a complex disease to help reveal its aetiology. Moreover, the gene features identified may have clinical applications as therapeutic targets.<\/jats:p>\n              <\/jats:sec>","DOI":"10.1186\/s12859-021-04454-4","type":"journal-article","created":{"date-parts":[[2021,11,24]],"date-time":"2021-11-24T11:02:59Z","timestamp":1637751779000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":12,"title":["Novel deep learning-based solution for identification of prognostic subgroups in liver cancer (Hepatocellular carcinoma)"],"prefix":"10.1186","volume":"22","author":[{"given":"Alice R.","family":"Owens","sequence":"first","affiliation":[]},{"given":"Caitr\u00edona E.","family":"McInerney","sequence":"additional","affiliation":[]},{"given":"Kevin M.","family":"Prise","sequence":"additional","affiliation":[]},{"given":"Darragh G.","family":"McArt","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1002-5079","authenticated-orcid":false,"given":"Anna","family":"Jurek-Loughrey","sequence":"additional","affiliation":[]}],"member":"297","published-online":{"date-parts":[[2021,11,24]]},"reference":[{"issue":"1","key":"4454_CR1","doi-asserted-by":"publisher","first-page":"182","DOI":"10.1016\/j.jhep.2018.03.019","volume":"69","author":"European Association For The Study Of The Liver","year":"2018","unstructured":"European Association For The Study Of The Liver. EASL clinical practice guidelines: management of hepatocellular carcinoma. J Hepatol. 2018;69(1):182\u2013236.","journal-title":"J Hepatol."},{"issue":"6","key":"4454_CR2","doi-asserted-by":"publisher","first-page":"1264","DOI":"10.1053\/j.gastro.2011.12.061","volume":"142","author":"HB El-Serag","year":"2012","unstructured":"El-Serag HB. Epidemiology of viral hepatitis and hepatocellular carcinoma. Gastroenterology. 2012;142(6):1264-1273.e1.","journal-title":"Gastroenterology"},{"key":"4454_CR3","doi-asserted-by":"publisher","first-page":"1","DOI":"10.4103\/jcar.JCar_9_16","volume":"16","author":"YA Ghouri","year":"2017","unstructured":"Ghouri YA, Mian I, Rowe JH. Review of hepatocellular carcinoma: Epidemiology, etiology, and carcinogenesis. J Carcinog. 2017;16:1.","journal-title":"J Carcinog."},{"key":"4454_CR4","first-page":"197","volume":"2017","author":"OB Poirion","year":"2018","unstructured":"Poirion OB, Chaudhary K, Garmire LX. Deep Learning data integration for better risk stratification models of bladder cancer. AMIA Jt Summits Transl Sci. 2018;2017:197\u2013206.","journal-title":"AMIA Jt Summits Transl Sci"},{"key":"4454_CR5","doi-asserted-by":"publisher","first-page":"477","DOI":"10.3389\/fgene.2018.00477","volume":"9","author":"L Zhang","year":"2018","unstructured":"Zhang L, Lv C, Jin Y, Cheng G, Fu Y, Yuan D, et al. Deep learning-based multi-omics data integration reveals two prognostic subtypes in high-risk neuroblastoma. Front Genet. 2018;9:477. https:\/\/doi.org\/10.3389\/fgene.2018.00477.","journal-title":"Front Genet"},{"issue":"6","key":"4454_CR6","doi-asserted-by":"publisher","first-page":"1248","DOI":"10.1158\/1078-0432.CCR-17-0853","volume":"24","author":"K Chaudhary","year":"2018","unstructured":"Chaudhary K, Poirion OB, Lu L, Garmire LX. Deep learning-based multi-omics integration robustly predicts survival in liver cancer. Clin Cancer Res. 2018;24(6):1248\u201359.","journal-title":"Clin Cancer Res"},{"issue":"1","key":"4454_CR7","doi-asserted-by":"publisher","first-page":"e00025-15","DOI":"10.1128\/mSystems.00025-15","volume":"1","author":"J Tan","year":"2016","unstructured":"Tan J, Hammond JH, Hogan DA, Greene CS. ADAGE-based integration of publicly available pseudomonas aeruginosa gene expression data with denoising autoencoders illuminates microbe-host interactions. MSystems. 2016;1(1):e00025-15.","journal-title":"MSystems"},{"issue":"1","key":"4454_CR8","doi-asserted-by":"publisher","first-page":"24","DOI":"10.1186\/s12874-018-0482-1","volume":"18","author":"JL Katzman","year":"2018","unstructured":"Katzman JL, Shaham U, Cloninger A, Bates J, Jiang T, Kluger Y. DeepSurv: personalized treatment recommender system using a Cox proportional hazards deep neural network. BMC Med Res Methodol. 2018;18(1):24. https:\/\/doi.org\/10.1186\/s12874-018-0482-1.","journal-title":"BMC Med Res Methodol"},{"issue":"2","key":"4454_CR9","doi-asserted-by":"publisher","first-page":"95","DOI":"10.1038\/s42256-019-0019-2","volume":"1","author":"GA Bello","year":"2019","unstructured":"Bello GA, Dawes TJW, Duan J, Biffi C, De Marvao A, Howard LSGE, et al. Deep-learning cardiac motion analysis for human survival prediction. Nat Mach Intell. 2019;1(2):95\u2013104.","journal-title":"Nat Mach Intell"},{"key":"4454_CR10","unstructured":"Yang B, Fu X, Sidiropoulos ND, Hong M. Towards k-means-friendly spaces: Simultaneous deep learning and clustering. In: international conference on machine learning. PMLR; 2017. p. 3861\u201370."},{"issue":"9","key":"4454_CR11","doi-asserted-by":"publisher","first-page":"1615","DOI":"10.1093\/bioinformatics\/btx812","volume":"34","author":"L Wei","year":"2018","unstructured":"Wei L, Jin Z, Yang S, Xu Y, Zhu Y, Ji Y. TCGA-assembler 2: software pipeline for retrieval and processing of TCGA\/CPTAC data. Bioinformatics. 2018;34(9):1615\u20137.","journal-title":"Bioinformatics"},{"key":"4454_CR12","unstructured":"Hastie T, Tibshirani R, Narasimhan B, Chu G. impute: impute: Imputation for microarray data. 2018."},{"issue":"5","key":"4454_CR13","doi-asserted-by":"publisher","first-page":"500","DOI":"10.1038\/ng.3547","volume":"48","author":"A Fujimoto","year":"2016","unstructured":"Fujimoto A, Furuta M, Totoki Y, Tsunoda T, Kato M, Shiraishi Y, Tanaka H, Taniguchi H, Kawakami Y, Ueno M, Gotoh K. Whole-genome mutational landscape and characterization of noncoding and structural mutations in liver cancer. Nat Genet. 2016;48(5):500\u20139.","journal-title":"Nat Genet"},{"key":"4454_CR14","unstructured":"Chollet F. Keras. 2015. Available from: https:\/\/keras.io"},{"key":"4454_CR15","unstructured":"Abadi M, Agarwal A, Barham P, Brevdo E, Chen Z, Citro C, et al. TensorFlow: large-scale machine learning on heterogeneous systems. 2015."},{"issue":"Oct","key":"4454_CR16","first-page":"2825","volume":"12","author":"F Pedregosa","year":"2011","unstructured":"Pedregosa F, Varoquaux G, Gramfort A, Michel V, Thirion B, Grisel O, et al. Scikit-learn: Machine learning in Python. J Mach Learn Res. 2011;12(Oct):2825\u201330.","journal-title":"J Mach Learn Res."},{"key":"4454_CR17","doi-asserted-by":"publisher","first-page":"53","DOI":"10.1016\/0377-0427(87)90125-7","volume":"20","author":"PJ Rousseeuw","year":"1987","unstructured":"Rousseeuw PJ. Silhouettes: a graphical aid to the interpretation and validation of cluster analysis. J Comput Appl Math. 1987;20:53\u201365.","journal-title":"J Comput Appl Math"},{"key":"4454_CR18","unstructured":"Therneau TM. A package for survival analysis in R. 2020. Available from: https:\/\/cran.r-project.org\/package=survival"},{"issue":"40","key":"4454_CR19","doi-asserted-by":"publisher","first-page":"1317","DOI":"10.21105\/joss.01317","volume":"4","author":"C Davidson-Pilon","year":"2019","unstructured":"Davidson-Pilon C. Lifelines: survival analysis in Python. J Open Source Softw. 2019;4(40):1317.","journal-title":"J Open Source Softw"},{"key":"4454_CR20","doi-asserted-by":"publisher","first-page":"352","DOI":"10.1038\/s41592-020-0772-5","volume":"17","author":"P Virtanen","year":"2020","unstructured":"Virtanen P, Gommers R, Oliphant TE, Haberland M, Reddy T, Cournapeau D, et al. SciPy 1.0: fundamental algorithms for scientific computing in python. Nat Methods. 2020;17:352.","journal-title":"Nat Methods"},{"issue":"1","key":"4454_CR21","doi-asserted-by":"publisher","first-page":"44","DOI":"10.1038\/nprot.2008.211","volume":"4","author":"BT Sherman","year":"2009","unstructured":"Sherman BT, Lempicki RA, et al. Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nat Protoc. 2009;4(1):44.","journal-title":"Nat Protoc."},{"key":"4454_CR22","doi-asserted-by":"publisher","unstructured":"Waskom M, Botvinnik O, O\u2019Kane D, Hobson P, Lukauskas S, Gemperline DC, et al. mwaskom\/seaborn: v0.8.1 (September 2017). Zenodo; 2017. https:\/\/doi.org\/10.5281\/zenodo.883859.","DOI":"10.5281\/zenodo.883859"},{"issue":"24","key":"4454_CR23","doi-asserted-by":"publisher","first-page":"7251","DOI":"10.7150\/thno.31155","volume":"9","author":"J Long","year":"2019","unstructured":"Long J, Chen P, Lin J, Bai Y, Yang X, Bian J, et al. DNA methylation-driven genes for constructing diagnostic, prognostic, and recurrence models for hepatocellular carcinoma. Theranostics. 2019;9(24):7251\u201367.","journal-title":"Theranostics"},{"key":"4454_CR24","doi-asserted-by":"publisher","first-page":"6079","DOI":"10.2147\/CMAR.S181396","volume":"10","author":"Y Zheng","year":"2018","unstructured":"Zheng Y, Liu Y, Zhao S, Zheng Z, Shen C, An L, et al. Large-scale analysis reveals a novel risk score to predict overall survival in hepatocellular carcinoma. Cancer Manag Res. 2018;10:6079\u201396.","journal-title":"Cancer Manag Res"},{"issue":"2","key":"4454_CR25","doi-asserted-by":"publisher","first-page":"319","DOI":"10.1002\/ijc.25336","volume":"128","author":"F Lan","year":"2011","unstructured":"Lan F, Wang H, Chen Y, Chan C, Ng SS, Li K, et al. Has-let-7g inhibits proliferation of hepatocellular carcinoma cells by downregulation of c-Myc and upregulation of p16INK4A. Int J Cancer. 2011;128(2):319\u201331.","journal-title":"Int J Cancer"},{"issue":"15","key":"4454_CR26","doi-asserted-by":"publisher","first-page":"2835","DOI":"10.7150\/ijbs.46285","volume":"16","author":"X Dong","year":"2020","unstructured":"Dong X, Feng M, Yang H, Liu H, Guo H, Gao X, et al. Rictor promotes cell migration and actin polymerization through regulating ABLIM1 phosphorylation in Hepatocellular Carcinoma. Int J Biol Sci. 2020;16(15):2835\u201352.","journal-title":"Int J Biol Sci."}],"container-title":["BMC Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s12859-021-04454-4.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1186\/s12859-021-04454-4\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s12859-021-04454-4.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2021,11,24]],"date-time":"2021-11-24T11:04:41Z","timestamp":1637751881000},"score":1,"resource":{"primary":{"URL":"https:\/\/bmcbioinformatics.biomedcentral.com\/articles\/10.1186\/s12859-021-04454-4"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,11,24]]},"references-count":26,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2021,12]]}},"alternative-id":["4454"],"URL":"https:\/\/doi.org\/10.1186\/s12859-021-04454-4","relation":{},"ISSN":["1471-2105"],"issn-type":[{"value":"1471-2105","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,11,24]]},"assertion":[{"value":"6 January 2021","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"20 October 2021","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"24 November 2021","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"Not applicable.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethics approval and consent to participate"}},{"value":"Not applicable.","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Consent for publication"}},{"value":"The authors declare that they have no competing interests.","order":4,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}],"article-number":"563"}}