{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,26]],"date-time":"2026-08-26T03:13:59Z","timestamp":1787714039614,"version":"build-2784847793"},"reference-count":47,"publisher":"Oxford University Press (OUP)","issue":"5","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2006,3,1]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:p>Motivation: Genotype\u2013phenotype modeling problems are often overcomplete, or ill-posed, since the number of potential predictors\u2014genes, proteins, mutations and their interactions\u2014is large relative to the number of measured outcomes. Such datasets can still be used to train sparse parameter models that generalize accurately, by exerting a principle similar to Occam's Razor: When many possible theories can explain the observations, the most simple is most likely to be correct. We apply this philosophy to modeling the drug response of Type-1 Human Immunodeficiency Virus (HIV-1). Owing to the decreasing expense of genetic sequencing relative to in vitro phenotype testing, a statistical model that reliably predicts viral drug response from genetic data is an important tool in the selection of antiretroviral therapy (ART). The optimization techniques described will have application to many genotype\u2013phenotype modeling problems for the purpose of enhancing clinical decisions.<\/jats:p><jats:p>Results: We describe two regression techniques for predicting viral phenotype in response to ART from genetic sequence data. Both techniques employ convex optimization for the continuous subset selection of a sparse set of model parameters. The first technique, the least absolute shrinkage and selection operator, uses the l1 norm loss function to create a sparse linear model; the second, the support vector machine with radial basis kernel functions, uses the \u03b5-insensitive loss function to create a sparse non-linear model. The techniques are applied to predict the response of the HIV-1 virus to 10 reverse transcriptase inhibitor and 7 protease inhibitor drugs. The genetic data are derived from the HIV coding sequences for the reverse transcriptase and protease enzymes. When tested by cross-validation with actual laboratory measurements, these models predict drug response phenotype more accurately than models previously discussed in the literature, and other canonical techniques described here. Key features of the methods that enable this performance are the tendency to generate simple models where many of the parameters are zero, and the convexity of the cost function, which assures that we can find model parameters to globally minimize the cost function for a particular training dataset.<\/jats:p><jats:p>Availability: Results, tables and figures are available at<\/jats:p><jats:p>Contact: \u00a0mrabinowitz@genesecurity.net<\/jats:p><jats:p>Supplementary information: An Appendix to accompany this article is available at Bioinformatics online.<\/jats:p>","DOI":"10.1093\/bioinformatics\/btk011","type":"journal-article","created":{"date-parts":[[2005,12,21]],"date-time":"2005-12-21T02:18:10Z","timestamp":1135131490000},"page":"541-549","source":"Crossref","is-referenced-by-count":16,"title":["Accurate prediction of HIV-1 drug response from the reverse transcriptase and protease amino acid sequences using sparse models created by convex optimization"],"prefix":"10.1093","volume":"22","author":[{"given":"Matthew","family":"Rabinowitz","sequence":"first","affiliation":[{"name":"Gene Security Network 1 \u00a0 1 \u00a0 \u00a0 Palo Alto, CA, USA"},{"name":"Department of Engineering, Stanford University 2 \u00a0 2 \u00a0 \u00a0 Palo Alto, CA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lance","family":"Myers","sequence":"additional","affiliation":[{"name":"Northwestern University School of Medicine 3 \u00a0 3 \u00a0 \u00a0 Chicago, IL, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Milena","family":"Banjevic","sequence":"additional","affiliation":[{"name":"Gene Security Network 1 \u00a0 1 \u00a0 \u00a0 Palo Alto, CA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Albert","family":"Chan","sequence":"additional","affiliation":[{"name":"Gene Security Network 1 \u00a0 1 \u00a0 \u00a0 Palo Alto, CA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Joshua","family":"Sweetkind-Singer","sequence":"additional","affiliation":[{"name":"Gene Security Network 1 \u00a0 1 \u00a0 \u00a0 Palo Alto, CA, USA"},{"name":"Department of Engineering, Stanford University 2 \u00a0 2 \u00a0 \u00a0 Palo Alto, CA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jessica","family":"Haberer","sequence":"additional","affiliation":[{"name":"Gene Security Network 1 \u00a0 1 \u00a0 \u00a0 Palo Alto, CA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kelly","family":"McCann","sequence":"additional","affiliation":[{"name":"Gene Security Network 1 \u00a0 1 \u00a0 \u00a0 Palo Alto, CA, USA"},{"name":"Department of Microbiology and Immunology, Stanford University Medical Center 4 \u00a0 4 \u00a0 \u00a0 Palo Alto, CA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Roland","family":"Wolkowicz","sequence":"additional","affiliation":[{"name":"Gene Security Network 1 \u00a0 1 \u00a0 \u00a0 Palo Alto, CA, USA"},{"name":"Department of Microbiology and Immunology, Stanford University Medical Center 4 \u00a0 4 \u00a0 \u00a0 Palo Alto, CA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"286","published-online":{"date-parts":[[2005,12,20]]},"reference":[{"key":"2023012408534149000_b1","article-title":"Agence Nationale de Recherches sur le SIDA, group AC11","author":"ANRS","year":"2004"},{"key":"2023012408534149000_b2","doi-asserted-by":"crossref","first-page":"3850","DOI":"10.1093\/nar\/gkg575","article-title":"Geno2pheno: Estimating phenotypic drug resistance from HIV-1 genotypes","volume":"31","author":"Beerenwinkel","year":"2003","journal-title":"Nucleic Acids Res."},{"key":"2023012408534149000_b3","doi-asserted-by":"crossref","first-page":"i16","DOI":"10.1093\/bioinformatics\/btg1001","article-title":"Methods for optimizing antiviral combination therapies","volume":"19","author":"Beerenwinkel","year":"2003","journal-title":"Bioinformatics"},{"key":"2023012408534149000_b4","doi-asserted-by":"crossref","first-page":"8271","DOI":"10.1073\/pnas.112177799","article-title":"Diversity and complexity of HIV-1 drug resistance: a bioinformatics approach to predicting phenotype from genotype","volume":"99","author":"Beerenwinkel","year":"2002","journal-title":"Proc. Natl Acad. Sci. USA"},{"key":"2023012408534149000_b5","doi-asserted-by":"crossref","DOI":"10.1017\/CBO9780511804441","volume-title":"Convex Optimization","author":"Boyd","year":"2004"},{"key":"2023012408534149000_b6","volume-title":"Classification and regression trees.","author":"Breiman","year":"1984"},{"key":"2023012408534149000_b7","volume-title":"A Tutorial on Support Vector Machines for Pattern Recognition","author":"Burges","year":"1998"},{"key":"2023012408534149000_b8","doi-asserted-by":"crossref","first-page":"381","DOI":"10.1001\/jama.283.3.381","article-title":"Antiretroviral therapy in adults: updated recommendations of the International AIDS Society-USA Panel","volume":"283","author":"Carpenter","year":"2000","journal-title":"JAMA"},{"key":"2023012408534149000_b9","doi-asserted-by":"crossref","DOI":"10.1111\/j.0006-341X.2001.01173.x","article-title":"Shrinkage estimators for covariance matrices","author":"Daniels","year":"2001"},{"key":"2023012408534149000_b10","first-page":"92","article-title":"Drug resistance mutations in HIV-1","volume":"11","author":"D'Aquila","year":"2003","journal-title":"Top HIV Med."},{"key":"2023012408534149000_b11","doi-asserted-by":"crossref","first-page":"1934","DOI":"10.1086\/375355","article-title":"Variable prediction of antiretroviral treatment outcome by different systems for interpreting genotypic human immunodeficiency virus type 1 drug resistance","volume":"187","author":"De Luca","year":"2003","journal-title":"J. Infect. Dis."},{"key":"2023012408534149000_b12","doi-asserted-by":"crossref","first-page":"583","DOI":"10.1177\/135965350400900406","article-title":"Construction, training and clinical validation of an interpretation system for genotypic HIV-1 drug resistance based on fuzzy rules revised by virologic outcomes","volume":"9","author":"De Luca","year":"2004","journal-title":"Antivir. Ther."},{"key":"2023012408534149000_b13","article-title":"Guidelines for the Use of Antiretroviral Agents in HIV-1-Infected Adults and Adolescents","author":"DHHS","year":"2004"},{"key":"2023012408534149000_b14","doi-asserted-by":"crossref","first-page":"2785","DOI":"10.1002\/sim.1516","article-title":"Non-parametric methods to predict HIV drug susceptibility phenotype from genotype","volume":"22","author":"DiRienzo","year":"2003","journal-title":"Stat. Med."},{"key":"2023012408534149000_b15","doi-asserted-by":"crossref","first-page":"98","DOI":"10.1093\/bioinformatics\/19.1.98","article-title":"Predicting HIV drug resistance with neural networks","volume":"19","author":"Draghici","year":"2003","journal-title":"Bioinformatics"},{"key":"2023012408534149000_b16","first-page":"407","article-title":"Least Angle Regression","volume":"32","author":"Efron","year":"2003","journal-title":"Ann. Stat."},{"key":"2023012408534149000_b17","article-title":"The Use of Log-Determinant Heuristics to Solve Rank Minimization Problems for Matrices","author":"Fazel","year":"2002"},{"key":"2023012408534149000_b18","article-title":"High-level dual (NRTI and PI) and triple class multidrug resistance (MDR) in a large health maintenance organization: prevalence, risk factors, and response to salvage therapy [abstract H-915]","author":"Fessel","year":"2003"},{"key":"2023012408534149000_b19","volume-title":"Lectures on the Cauchy Problem in Linear Partial Differential Equations","author":"Hadamard","year":"1923"},{"key":"2023012408534149000_b20","doi-asserted-by":"crossref","DOI":"10.1007\/978-0-387-21606-5","volume-title":"Elements of Statistical Learning, Data Mining, and Prediction","author":"Hastie","year":"2001"},{"key":"2023012408534149000_b21","volume-title":"General Additive Models","author":"Hastie","year":"1996"},{"key":"2023012408534149000_b22","article-title":"The HIV RT and Protease Sequence Database at Stanford University","author":"HIVDB","year":"2004"},{"key":"2023012408534149000_b23","volume-title":"Principal Component Analysis","author":"Jolliffe","year":"2002"},{"key":"2023012408534149000_b24","doi-asserted-by":"crossref","first-page":"72","DOI":"10.1046\/j.1468-1293.2003.00131.x","article-title":"Discrepant results in the interpretation of HIV-1 drug-resistance genotypic data among widely used algorithms","volume":"4","author":"Kijak","year":"2003","journal-title":"HIV Med."},{"key":"2023012408534149000_b25","doi-asserted-by":"crossref","first-page":"S626","DOI":"10.1002\/gepi.2001.21.s1.s626","article-title":"Sequence analysis using logic regression","volume":"21","author":"Kooperberg","year":"2001","journal-title":"Genet Epidemiol."},{"key":"2023012408534149000_b26","doi-asserted-by":"crossref","first-page":"975","DOI":"10.1097\/00002030-199610090-00007","article-title":"Drug resistance and virologic response in NUCA 3001, a randomized trial of lamivudine (3TC) versus zidovudine (ZDV) versus ZDV plus 3TC in previously untreated patients","volume":"10","author":"Kuritzkes","year":"1996","journal-title":"AIDS"},{"key":"2023012408534149000_b27","volume-title":"Multivariate Analysis","author":"Mardia","year":"1980"},{"key":"2023012408534149000_b28","doi-asserted-by":"crossref","first-page":"727","DOI":"10.1097\/00002030-200203290-00008","article-title":"Phenotypic or genotypic resistance testing for choosing antiretroviral therapy after treatment failure: a randomized trial","volume":"16","author":"Meynard","year":"2002","journal-title":"AIDS"},{"key":"2023012408534149000_b29","doi-asserted-by":"crossref","first-page":"760","DOI":"10.1038\/nm0796-760","article-title":"Ordered accumulation of mutations in HIV protease confers resistance to ritonavir","volume":"2","author":"Molla","year":"1996","journal-title":"Nat. Med."},{"key":"2023012408534149000_b30","volume-title":"Practical Optimization","author":"Murray","year":"1981"},{"key":"2023012408534149000_b31","volume-title":"Applied Linear Statistical Models: Regression, Analysis of Variance and Experimental Designs","author":"Neter","year":"1990","edition":"3rd edn"},{"key":"2023012408534149000_b32","doi-asserted-by":"crossref","first-page":"565","DOI":"10.1038\/ng1369","article-title":"Human genes that limit AIDS","volume":"36","author":"O'Brien","year":"2004","journal-title":"Nat. Genet."},{"key":"2023012408534149000_b33","doi-asserted-by":"crossref","first-page":"389","DOI":"10.1093\/imanum\/20.3.389","article-title":"A new approach to variable selection in least squares problems","volume":"20","author":"Osborne","year":"2000","journal-title":"IMA J. Numer. Anal."},{"key":"2023012408534149000_b34","doi-asserted-by":"crossref","first-page":"8","DOI":"10.1097\/00126334-200305010-00002","article-title":"HIV-1 protease and reverse transcriptase mutation patterns responsible for discordances between genotypic drug resistance interpretation algorithms","volume":"33","author":"Ravela","year":"2003","journal-title":"J. Acquir. Immune Defic. Syndr."},{"key":"2023012408534149000_b35","doi-asserted-by":"crossref","first-page":"298","DOI":"10.1093\/nar\/gkg100","article-title":"Human immunodeficiency virus reverse transcriptase and protease sequence database","volume":"31","author":"Rhee","year":"2003","journal-title":"Nucleic Acids Res."},{"key":"2023012408534149000_b36","volume-title":"Modern Regression Methods","author":"Ryan","year":"1996"},{"key":"2023012408534149000_b37","volume-title":"Advances in Kernel Methods\u2014Support Vector Learning","author":"Sch\u00f6lkopf","year":"1999"},{"key":"2023012408534149000_b38","doi-asserted-by":"crossref","first-page":"247","DOI":"10.1128\/CMR.15.2.247-277.2002","article-title":"Genotypic testing for human immunodeficiency virus type 1 drug resistance","volume":"15","author":"Shafer","year":"2002","journal-title":"Clin. Microbiol. Rev."},{"key":"2023012408534149000_b39","volume-title":"NIAID ACTG Data Analysis Concept Sheet Proposal: Genotype predictors of clinical outcome across completed ACTG clinical trials","author":"Shafer","year":"2004"},{"key":"2023012408534149000_b40","doi-asserted-by":"crossref","first-page":"373","DOI":"10.1080\/09540120050123765","article-title":"Adherence\u2014sticking to the evidence","volume":"12","author":"Sherr","year":"2000","journal-title":"AIDS Care"},{"key":"2023012408534149000_b41","article-title":"Log-Penalized Linear Regression","author":"Singer","year":"2004"},{"key":"2023012408534149000_b42","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 Royal Statist. Soc. B"},{"key":"2023012408534149000_b43","article-title":"UNAIDS\/WHO. Report on the Global AIDS epidemic. Geneva, Switzerland","author":"UNAIDS","year":"2004"},{"key":"2023012408534149000_b44","volume-title":"The Nature of Statistical Learning Theory","author":"Vapnik","year":"1998"},{"key":"2023012408534149000_b45","doi-asserted-by":"crossref","first-page":"653","DOI":"10.1086\/377453","article-title":"Enhanced prediction of lopinavir resistancefrom genotype by use of artificial neural networks","volume":"188","author":"Wang","year":"2003","journal-title":"J. Infect Dis."},{"key":"2023012408534149000_b46","doi-asserted-by":"crossref","first-page":"343","DOI":"10.1177\/135965350400900307","article-title":"Simple linear model provides highly accurate genotypic predictions of HIV-1 drug resistance","volume":"9","author":"Wang","year":"2004","journal-title":"Antivir Ther."},{"key":"2023012408534149000_b47","doi-asserted-by":"crossref","first-page":"813","DOI":"10.7326\/0003-4819-131-11-199912070-00003","article-title":"HIV-1 genotypic resistance patterns predict response to saquinavir-ritonavir therapy in patients in whom previous protease inhibitor therapy had failed","volume":"131","author":"Zolopa","year":"1999","journal-title":"Ann. Intern. Med."}],"container-title":["Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/22\/5\/541\/48839863\/bioinformatics_22_5_541.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/22\/5\/541\/48839863\/bioinformatics_22_5_541.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,1,6]],"date-time":"2025-01-06T17:30:24Z","timestamp":1736184624000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article\/22\/5\/541\/205390"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2005,12,20]]},"references-count":47,"journal-issue":{"issue":"5","published-print":{"date-parts":[[2006,3,1]]}},"URL":"https:\/\/doi.org\/10.1093\/bioinformatics\/btk011","relation":{},"ISSN":["1367-4811","1367-4803"],"issn-type":[{"value":"1367-4811","type":"electronic"},{"value":"1367-4803","type":"print"}],"subject":[],"published-other":{"date-parts":[[2006,3,1]]},"published":{"date-parts":[[2005,12,20]]}}}