{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,19]],"date-time":"2026-05-19T23:48:20Z","timestamp":1779234500030,"version":"3.51.4"},"reference-count":44,"publisher":"Oxford University Press (OUP)","issue":"7","funder":[{"DOI":"10.13039\/100000002","name":"NIH","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100000002","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000066","name":"National Institute of Environmental Health Sciences","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100000066","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2021,5,17]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:sec><jats:title>Motivation<\/jats:title><jats:p>The recently proposed knockoff filter is a general framework for controlling the false discovery rate (FDR) when performing variable selection. This powerful new approach generates a \u2018knockoff\u2019 of each variable tested for exact FDR control. Imitation variables that mimic the correlation structure found within the original variables serve as negative controls for statistical inference. Current applications of knockoff methods use linear regression models and conduct variable selection only for variables existing in model functions. Here, we extend the use of knockoffs for machine learning with boosted trees, which are successful and widely used in problems where no prior knowledge of model function is required. However, currently available importance scores in tree models are insufficient for variable selection with FDR control.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>We propose a novel strategy for conducting variable selection without prior model topology knowledge using the knockoff method with boosted tree models. We extend the current knockoff method to model-free variable selection through the use of tree-based models. Additionally, we propose and evaluate two new sampling methods for generating knockoffs, namely the sparse covariance and principal component knockoff methods. We test and compare these methods with the original knockoff method regarding their ability to control type I errors and power. In simulation tests, we compare the properties and performance of importance test statistics of tree models. The results include different combinations of knockoffs and importance test statistics. We consider scenarios that include main-effect, interaction, exponential and second-order models while assuming the true model structures are unknown. We apply our algorithm for tumor purity estimation and tumor classification using Cancer Genome Atlas (TCGA) gene expression data. Our results show improved discrimination between difficult-to-discriminate cancer types.<\/jats:p><\/jats:sec><jats:sec><jats:title>Availability and implementation<\/jats:title><jats:p>The proposed algorithm is included in the KOBT package, which is available at https:\/\/cran.r-project.org\/web\/packages\/KOBT\/index.html.<\/jats:p><\/jats:sec><jats:sec><jats:title>Supplementary information<\/jats:title><jats:p>Supplementary data are available at Bioinformatics online.<\/jats:p><\/jats:sec>","DOI":"10.1093\/bioinformatics\/btaa770","type":"journal-article","created":{"date-parts":[[2020,9,10]],"date-time":"2020-09-10T13:56:27Z","timestamp":1599746187000},"page":"976-983","source":"Crossref","is-referenced-by-count":18,"title":["Knockoff boosted tree for model-free variable selection"],"prefix":"10.1093","volume":"37","author":[{"given":"Tao","family":"Jiang","sequence":"first","affiliation":[{"name":"Department of Statistics, Bioinformatics Research Center, North Carolina State University , Raleigh, NC 27695, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yuanyuan","family":"Li","sequence":"additional","affiliation":[{"name":"Biostatistics & Computational Biology Branch, National Institute of Environmental Health Sciences, Durham , NC 27709, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1346-2493","authenticated-orcid":false,"given":"Alison A","family":"Motsinger-Reif","sequence":"additional","affiliation":[{"name":"Biostatistics & Computational Biology Branch, National Institute of Environmental Health Sciences, Durham , NC 27709, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"286","published-online":{"date-parts":[[2020,9,23]]},"reference":[{"key":"2023051612171777800_btaa770-B1","doi-asserted-by":"crossref","first-page":"8971","DOI":"10.1038\/ncomms9971","article-title":"Systematic pan-cancer analysis of tumour purity","volume":"6","author":"Aran","year":"2015","journal-title":"Nat. Commun"},{"key":"2023051612171777800_btaa770-B2","doi-asserted-by":"crossref","first-page":"2055","DOI":"10.1214\/15-AOS1337","article-title":"Controlling the false discovery rate via knockoffs","volume":"43","author":"Barber","year":"2015","journal-title":"Ann. Stat"},{"key":"2023051612171777800_btaa770-B3","doi-asserted-by":"crossref","first-page":"811","DOI":"10.1007\/s00109-014-1181-y","article-title":"Hox genes and their role in the development of human cancers","volume":"92","author":"Bhatlekar","year":"2014","journal-title":"J. Mol. Med"},{"key":"2023051612171777800_btaa770-B4","doi-asserted-by":"crossref","first-page":"807","DOI":"10.1093\/biomet\/asr054","article-title":"Sparse estimation of a covariance matrix","volume":"98","author":"Bien","year":"2011","journal-title":"Biometrika"},{"key":"2023051612171777800_btaa770-B5","doi-asserted-by":"crossref","first-page":"5","DOI":"10.1023\/A:1010933404324","article-title":"Random forests","volume":"45","author":"Breiman","year":"2001","journal-title":"Mach. Learn"},{"key":"2023051612171777800_btaa770-B6","doi-asserted-by":"crossref","first-page":"551","DOI":"10.1111\/rssb.12265","article-title":"Panning for gold: \u2018model-x\u2019 knockoffs for high dimensional controlled variable selection","volume":"80","author":"Candes","year":"2018","journal-title":"J. R. Stat. Soc. Ser. B (Stat. Methodol.)"},{"key":"2023051612171777800_btaa770-B7","doi-asserted-by":"crossref","first-page":"413","DOI":"10.1038\/nbt.2203","article-title":"Absolute quantification of somatic dna alterations in human cancer","volume":"30","author":"Carter","year":"2012","journal-title":"Nat. Biotechnol"},{"key":"2023051612171777800_btaa770-B8","first-page":"785","author":"Chen","year":"2016"},{"key":"2023051612171777800_btaa770-B9","doi-asserted-by":"crossref","first-page":"484","DOI":"10.1214\/13-AOS1090","article-title":"Exact and asymptotically robust permutation tests","volume":"41","author":"Chung","year":"2013","journal-title":"Ann. Stat"},{"key":"2023051612171777800_btaa770-B10","first-page":"289","author":"Frid-Adar","year":"2018"},{"key":"2023051612171777800_btaa770-B11","doi-asserted-by":"crossref","first-page":"337","DOI":"10.1214\/aos\/1016218223","article-title":"Additive logistic regression: a statistical view of boosting (with discussion and a rejoinder by the authors)","volume":"28","author":"Friedman","year":"2000","journal-title":"Ann. Stat"},{"key":"2023051612171777800_btaa770-B12","doi-asserted-by":"crossref","first-page":"1189","DOI":"10.1214\/aos\/1013203451","article-title":"Greedy function approximation: a gradient boosting machine","volume":"29","author":"Friedman","year":"2001","journal-title":"Ann. Stat"},{"key":"2023051612171777800_btaa770-B13","doi-asserted-by":"crossref","first-page":"531","DOI":"10.1126\/science.286.5439.531","article-title":"Molecular classification of cancer: class discovery and class prediction by gene expression monitoring","volume":"286","author":"Golub","year":"1999","journal-title":"Science"},{"key":"2023051612171777800_btaa770-B14","first-page":"513","article-title":"A kernel method for the two-sample-problem","author":"Gretton","year":"2007","journal-title":"Adv. Neural Inf. Process. Syst"},{"key":"2023051612171777800_btaa770-B15","doi-asserted-by":"crossref","first-page":"83","DOI":"10.1007\/BF02985802","article-title":"The elements of statistical learning: data mining, inference and prediction","volume":"27","author":"Hastie","year":"2005","journal-title":"Math. Intell"},{"key":"2023051612171777800_btaa770-B16","doi-asserted-by":"crossref","first-page":"291","DOI":"10.1016\/j.cell.2018.03.022","article-title":"Cell-of-origin patterns dominate the molecular classification of 10,000 tumors from 33 types of cancer","volume":"173","author":"Hoadley","year":"2018","journal-title":"Cell"},{"key":"2023051612171777800_btaa770-B17","doi-asserted-by":"crossref","first-page":"86","DOI":"10.1186\/1471-2105-13-86","article-title":"Dna methylation arrays as surrogate measures of cell mixture distribution","volume":"13","author":"Houseman","year":"2012","journal-title":"BMC Bioinformatics"},{"key":"2023051612171777800_btaa770-B18","doi-asserted-by":"crossref","first-page":"611","DOI":"10.1016\/j.devcel.2005.01.015","article-title":"The stomach mesenchymal transcription factor Barx1 specifies gastric epithelial identity through inhibition of transient wnt signaling","volume":"8","author":"Kim","year":"2005","journal-title":"Dev. Cell"},{"key":"2023051612171777800_btaa770-B19","doi-asserted-by":"crossref","first-page":"1081","DOI":"10.1242\/dev.056317","article-title":"Regulation of mouse stomach development and Barx1 expression by specific micrornas","volume":"138","author":"Kim","year":"2011","journal-title":"Development"},{"key":"2023051612171777800_btaa770-B20","doi-asserted-by":"crossref","first-page":"589","DOI":"10.1198\/016214501753168271","article-title":"Classification trees with unbiased multiway splits","volume":"96","author":"Kim","year":"2001","journal-title":"J. Am. Stat. Assoc"},{"key":"2023051612171777800_btaa770-B21","author":"LeCun","year":"1995"},{"key":"2023051612171777800_btaa770-B22","doi-asserted-by":"crossref","first-page":"285","DOI":"10.1111\/j.1467-9868.2005.00502.x","article-title":"Model-free variable selection","volume":"67","author":"Li","year":"2005","journal-title":"J. R. Stat. Soc. Ser. B (Stat. Methodol.)"},{"key":"2023051612171777800_btaa770-B23","doi-asserted-by":"crossref","first-page":"508","DOI":"10.1186\/s12864-017-3906-0","article-title":"A comprehensive genomic pan-cancer classification using the cancer genome atlas gene expression data","volume":"18","author":"Li","year":"2017","journal-title":"BMC Genomics"},{"key":"2023051612171777800_btaa770-B24","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1186\/s12864-019-6412-8","article-title":"Putative biomarkers for predicting tumor sample purity based on gene expression data","volume":"20","author":"Li","year":"2019","journal-title":"BMC Genomics"},{"key":"2023051612171777800_btaa770-B25","doi-asserted-by":"crossref","first-page":"1114","DOI":"10.3389\/fonc.2019.01114","article-title":"Bioinformatic identification of miR-622 key target genes and experimental validation of the miR-622-RNF8 axis in breast cancer","volume":"9","author":"Liu","year":"2019","journal-title":"Front. Oncol"},{"key":"2023051612171777800_btaa770-B26","first-page":"361","article-title":"Regression tress with unbiased variable selection and interaction detection","volume":"12","author":"Loh","year":"2002","journal-title":"Stat. Sin"},{"key":"2023051612171777800_btaa770-B27","author":"Lundberg","year":"2018"},{"key":"2023051612171777800_btaa770-B28","author":"Lundberg","year":"2017"},{"key":"2023051612171777800_btaa770-B29","first-page":"4765","article-title":"A unified approach to interpreting model predictions","author":"Lundberg","year":"2017","journal-title":"Adv. Neural Inf. Process. Syst"},{"key":"2023051612171777800_btaa770-B30","doi-asserted-by":"crossref","first-page":"583","DOI":"10.1037\/met0000087","article-title":"Finding structure in data using multivariate tree boosting","volume":"21","author":"Miller","year":"2016","journal-title":"Psychol. Methods"},{"key":"2023051612171777800_btaa770-B31","author":"Nielsen","year":"2016"},{"key":"2023051612171777800_btaa770-B32","doi-asserted-by":"crossref","DOI":"10.2202\/1544-6115.1252","article-title":"Accurate ranking of differentially expressed genes by a distribution-free shrinkage approach","volume":"6","author":"Opgen-Rhein","year":"2007","journal-title":"Stat. Appl. Genet. Mol. Biol"},{"key":"2023051612171777800_btaa770-B33","doi-asserted-by":"crossref","first-page":"287","DOI":"10.1023\/A:1007618119488","article-title":"Soft margins for adaboost","volume":"42","author":"R\u00e4tsch","year":"2001","journal-title":"Mach. Learn"},{"key":"2023051612171777800_btaa770-B34","doi-asserted-by":"crossref","first-page":"6341","DOI":"10.1080\/00036846.2018.1489520","article-title":"Using regression tree ensembles to model interaction effects: a graphical approach","volume":"50","author":"Schiltz","year":"2018","journal-title":"Appl. Econ"},{"key":"2023051612171777800_btaa770-B35","doi-asserted-by":"crossref","first-page":"744","DOI":"10.3390\/cancers11060744","article-title":"False discovery rate control in cancer biomarker selection using knockoffs","volume":"11","author":"Shen","year":"2019","journal-title":"Cancers"},{"key":"2023051612171777800_btaa770-B36","first-page":"2951","article-title":"Practical bayesian optimization of machine learning algorithms","author":"Snoek","year":"2012","journal-title":"Adv. Neural Inf. Process. Syst"},{"key":"2023051612171777800_btaa770-B37","first-page":"130","article-title":"Decision tree methods: applications for classification and prediction","volume":"27","author":"Song","year":"2015","journal-title":"Shanghai Arch. Psychiatry"},{"key":"2023051612171777800_btaa770-B38","doi-asserted-by":"crossref","first-page":"457","DOI":"10.1111\/j.1467-9876.2010.00754.x","article-title":"Interaction trees: exploring the differential effects of an intervention programme for breast cancer survivors","volume":"60","author":"Su","year":"2011","journal-title":"J. R. Stat. Soc. Ser. C (Appl. Stat.)"},{"key":"2023051612171777800_btaa770-B39","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 (Methodol.)"},{"key":"2023051612171777800_btaa770-B40","doi-asserted-by":"crossref","first-page":"286","DOI":"10.1046\/j.1440-1827.2002.01353.x","article-title":"Heterogeneity of expression of cytokeratin subtypes in squamous cell carcinoma of the lung: with special reference to CK14 overexpression in cancer of high-proliferative and lymphogenous metastatic potential","volume":"52","author":"Tsubokawa","year":"2002","journal-title":"Pathol. Int"},{"key":"2023051612171777800_btaa770-B41","doi-asserted-by":"crossref","first-page":"669","DOI":"10.1038\/nri3902","article-title":"Immunological hallmarks of stromal cells in the tumour microenvironment","volume":"15","author":"Turley","year":"2015","journal-title":"Nat. Rev. Immunol"},{"key":"2023051612171777800_btaa770-B42","doi-asserted-by":"crossref","first-page":"2612","DOI":"10.1038\/ncomms3612","article-title":"Inferring tumour purity and stromal and immune cell admixture from expression data","volume":"4","author":"Yoshihara","year":"2013","journal-title":"Nat. Commun"},{"key":"2023051612171777800_btaa770-B43","doi-asserted-by":"crossref","first-page":"1538","DOI":"10.1214\/009053605000000255","article-title":"Boosting with early stopping: convergence and consistency","volume":"33","author":"Zhang","year":"2005","journal-title":"Ann. Stat"},{"key":"2023051612171777800_btaa770-B44","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 (Stat. Methodol.)"}],"container-title":["Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"http:\/\/academic.oup.com\/bioinformatics\/advance-article-pdf\/doi\/10.1093\/bioinformatics\/btaa770\/35334878\/btaa770.pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/37\/7\/976\/50341175\/btaa770.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/37\/7\/976\/50341175\/btaa770.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,8,13]],"date-time":"2024-08-13T15:24:51Z","timestamp":1723562691000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article\/37\/7\/976\/5910548"}},"subtitle":[],"editor":[{"given":"Jonathan","family":"Wren","sequence":"additional","affiliation":[],"role":[{"role":"editor","vocabulary":"crossref"}]}],"short-title":[],"issued":{"date-parts":[[2020,9,23]]},"references-count":44,"journal-issue":{"issue":"7","published-print":{"date-parts":[[2021,5,17]]}},"URL":"https:\/\/doi.org\/10.1093\/bioinformatics\/btaa770","relation":{},"ISSN":["1367-4803","1367-4811"],"issn-type":[{"value":"1367-4803","type":"print"},{"value":"1367-4811","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2021,4,1]]},"published":{"date-parts":[[2020,9,23]]}}}