{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,3]],"date-time":"2026-03-03T20:12:44Z","timestamp":1772568764000,"version":"3.50.1"},"reference-count":22,"publisher":"Oxford University Press (OUP)","issue":"21","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2011,11,1]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:p>Motivation: The area under the receiver operating characteristic (ROC) curve (AUC), long regarded as a \u2018golden\u2019 measure for the predictiveness of a continuous score, has propelled the need to develop AUC-based predictors. However, the AUC-based ensemble methods are rather scant, largely due to the fact that the associated objective function is neither continuous nor concave. Indeed, there is no reliable numerical algorithm identifying optimal combination of a set of biomarkers to maximize the AUC, especially when the number of biomarkers is large.<\/jats:p><jats:p>Results: We have proposed a novel AUC-based statistical ensemble methods for combining multiple biomarkers to differentiate a binary response of interest. Specifically, we propose to replace the non-continuous and non-convex AUC objective function by a convex surrogate loss function, whose minimizer can be efficiently identified. With the established framework, the lasso and other regularization techniques enable feature selections. Extensive simulations have demonstrated the superiority of the new methods to the existing methods. The proposal has been applied to a gene expression dataset to construct gene expression scores to differentiate elderly women with low bone mineral density (BMD) and those with normal BMD. The AUCs of the resulting scores in the independent test dataset has been satisfactory.<\/jats:p><jats:p>Conclusion: Aiming for directly maximizing AUC, the proposed AUC-based ensemble method provides an efficient means of generating a stable combination of multiple biomarkers, which is especially useful under the high-dimensional settings.<\/jats:p><jats:p>Contact: \u00a0lutian@stanford.edu<\/jats:p><jats:p>Supplementary Information: \u00a0Supplementary data are available at Bioinformatics online.<\/jats:p>","DOI":"10.1093\/bioinformatics\/btr516","type":"journal-article","created":{"date-parts":[[2011,9,10]],"date-time":"2011-09-10T04:29:25Z","timestamp":1315628965000},"page":"3050-3055","source":"Crossref","is-referenced-by-count":15,"title":["AUC-based biomarker ensemble with an application on gene scores predicting low bone mineral density"],"prefix":"10.1093","volume":"27","author":[{"given":"X. G.","family":"Zhao","sequence":"first","affiliation":[{"name":"1 Department of Bone and Joint Surgery, The First Affiliated Hospital of Xi'an Medical University, Xi'an 710077, Shaanxi Province, P.R. China, 2Department of Biostatistics, Harvard University, Boston, MA 02115 and 3Department of Health Research and Policy, Stanford University, Palo Alto, CA 94301, USA"}]},{"given":"W.","family":"Dai","sequence":"additional","affiliation":[{"name":"1 Department of Bone and Joint Surgery, The First Affiliated Hospital of Xi'an Medical University, Xi'an 710077, Shaanxi Province, P.R. China, 2Department of Biostatistics, Harvard University, Boston, MA 02115 and 3Department of Health Research and Policy, Stanford University, Palo Alto, CA 94301, USA"}]},{"given":"Y.","family":"Li","sequence":"additional","affiliation":[{"name":"1 Department of Bone and Joint Surgery, The First Affiliated Hospital of Xi'an Medical University, Xi'an 710077, Shaanxi Province, P.R. China, 2Department of Biostatistics, Harvard University, Boston, MA 02115 and 3Department of Health Research and Policy, Stanford University, Palo Alto, CA 94301, USA"}]},{"given":"L.","family":"Tian","sequence":"additional","affiliation":[{"name":"1 Department of Bone and Joint Surgery, The First Affiliated Hospital of Xi'an Medical University, Xi'an 710077, Shaanxi Province, P.R. China, 2Department of Biostatistics, Harvard University, Boston, MA 02115 and 3Department of Health Research and Policy, Stanford University, Palo Alto, CA 94301, USA"}]}],"member":"286","published-online":{"date-parts":[[2011,9,9]]},"reference":[{"key":"2023012511341982500_B1","doi-asserted-by":"crossref","first-page":"216","DOI":"10.1093\/biostatistics\/kxm037","article-title":"Robust combination of multiple diagnostic tests for classifying censored event times","volume":"9","author":"Cai","year":"2008","journal-title":"Biostatistics"},{"key":"2023012511341982500_B2","doi-asserted-by":"crossref","first-page":"394","DOI":"10.1111\/j.1541-0420.2008.01074.x","article-title":"Regularized estimation for the accelerated failure time model","volume":"65","author":"Cai","year":"2009","journal-title":"Biometrics"},{"key":"2023012511341982500_B3","doi-asserted-by":"crossref","first-page":"12S","DOI":"10.1016\/S0002-9343(97)90022-X","article-title":"The crippling consequences of fractures and their impact on quality of life","volume":"103","author":"Cooper","year":"1997","journal-title":"Am. J. Med."},{"key":"2023012511341982500_B4","volume-title":"Gradient directed regularization for linear regression and classification.","author":"Friedman","year":"2004"},{"key":"2023012511341982500_B5","first-page":"1","article-title":"Regularization paths for generalized linear models via coordinate descent","volume":"33","author":"Friedman","year":"2010","journal-title":"J. Stat. Softwr."},{"key":"2023012511341982500_B6","doi-asserted-by":"crossref","first-page":"337","DOI":"10.1214\/aos\/1016218223","article-title":"Additive logistic regression: a statistical view of boosting","volume":"28","author":"Friendman","year":"2000","journal-title":"Ann. Stat."},{"key":"2023012511341982500_B7","first-page":"352","article-title":"Discussion of \u201csupport vector machines with applications\u201d by Javier Moguerza and Alberto Munoz","volume":"21","author":"Hastie","year":"2006","journal-title":"Stat. Sci."},{"key":"2023012511341982500_B8","doi-asserted-by":"crossref","first-page":"909","DOI":"10.1177\/0272989X08318462","article-title":"A procedure for determining whether a simple combination of diagnostic tests may be noninferior to the theoretical optimum combination","volume":"28","author":"Jin","year":"2008","journal-title":"Med. Decis Making"},{"key":"2023012511341982500_B9","doi-asserted-by":"crossref","first-page":"314","DOI":"10.1186\/1471-2105-11-314","article-title":"A boosting method for maximizing the partial area under the ROC curve","volume":"11","author":"Komori","year":"2010","journal-title":"BMC Bioinformatics"},{"key":"2023012511341982500_B10","doi-asserted-by":"crossref","first-page":"4356","DOI":"10.1093\/bioinformatics\/bti724","article-title":"Regularized roc method for disease classification and biomarker selection with microarray data","volume":"21","author":"Ma","year":"2005","journal-title":"Bioinformatics"},{"key":"2023012511341982500_B11","doi-asserted-by":"crossref","first-page":"751","DOI":"10.1111\/j.1541-0420.2006.00731.x","article-title":"Combining multiple markers for classification using ROC","volume":"63","author":"Ma","year":"2007","journal-title":"Biometrics"},{"key":"2023012511341982500_B12","doi-asserted-by":"crossref","DOI":"10.1093\/oso\/9780198509844.001.0001","volume-title":"The Statistical Evaluation of Medical Tests for Classification and Prediction.","author":"Pepe","year":"2003"},{"key":"2023012511341982500_B13","doi-asserted-by":"crossref","first-page":"221","DOI":"10.1111\/j.1541-0420.2005.00420.x","article-title":"Combining predictors for classification using the area under the receiver operating characteristic curve","volume":"62","author":"Pepe","year":"2006","journal-title":"Biometrics"},{"key":"2023012511341982500_B14","doi-asserted-by":"crossref","first-page":"604","DOI":"10.1016\/j.bone.2009.11.007","article-title":"Eight genes are highly associated with bmd variation in postmenopausal caucasian women","volume":"46","author":"Reppe","year":"2010","journal-title":"Bone"},{"key":"2023012511341982500_B15","doi-asserted-by":"crossref","first-page":"1012","DOI":"10.1214\/009053606000001370","article-title":"Piecewise linear regularized solution paths","volume":"35","author":"Rosset","year":"2007","journal-title":"Ann. Stat."},{"key":"2023012511341982500_B16","doi-asserted-by":"crossref","first-page":"267","DOI":"10.1111\/j.2517-6161.1996.tb02080.x","article-title":"Regression shrinkage and selection via the lasso","volume":"58","author":"Tibshirani","year":"1996","journal-title":"J. R. Stat. Soc., Ser. B"},{"key":"2023012511341982500_B17","first-page":"477","article-title":"On the analysis of glycomics mass spectrometry data via the regularized area under the ROC curve","volume":"8","author":"Ye","year":"2007","journal-title":"Bioinformatics"},{"key":"2023012511341982500_B18","doi-asserted-by":"crossref","DOI":"10.1002\/9780470317082","volume-title":"Statistical Methods in Diagnostic Medicine.","author":"Zhou","year":"2002"},{"key":"2023012511341982500_B19","article-title":"Variable selection using the optimal roc curve: An application to a traditional chinese medicine study on osteoporosis disease","author":"Zhou","year":"2011","journal-title":"Stat. Med."},{"key":"2023012511341982500_B20","doi-asserted-by":"crossref","first-page":"1418","DOI":"10.1198\/016214506000000735","article-title":"The adaptive lasso and its oracle properties","volume":"101","author":"Zou","year":"2006","journal-title":"J. Am. Stat. Assoc."},{"key":"2023012511341982500_B21","doi-asserted-by":"crossref","first-page":"301","DOI":"10.1111\/j.1467-9868.2005.00503.x","article-title":"Regularization and variable selection via the elastic net","volume":"67","author":"Zou","year":"2005","journal-title":"J. R. Stat. Soc. Ser. B"},{"key":"2023012511341982500_B22","first-page":"1509","article-title":"One-step sparse estimates in nonconcave penalized likelihood models","volume":"36","author":"Zou","year":"2008","journal-title":"Ann. Stat."}],"container-title":["Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/27\/21\/3050\/48862577\/bioinformatics_27_21_3050.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/27\/21\/3050\/48862577\/bioinformatics_27_21_3050.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,3,10]],"date-time":"2025-03-10T10:25:26Z","timestamp":1741602326000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article\/27\/21\/3050\/218320"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2011,9,9]]},"references-count":22,"journal-issue":{"issue":"21","published-print":{"date-parts":[[2011,11,1]]}},"URL":"https:\/\/doi.org\/10.1093\/bioinformatics\/btr516","relation":{},"ISSN":["1367-4811","1367-4803"],"issn-type":[{"value":"1367-4811","type":"electronic"},{"value":"1367-4803","type":"print"}],"subject":[],"published-other":{"date-parts":[[2011,11,1]]},"published":{"date-parts":[[2011,9,9]]}}}