Regularized estimation in sparse high-dimensional multivariate regression, with application to a DNA methylation study.

IF 0.8 4区数学 Q4 BIOCHEMISTRY & MOLECULAR BIOLOGY

Statistical Applications in Genetics and Molecular Biology Pub Date : 2017-07-26 DOI:10.1515/sagmb-2016-0073

Haixiang Zhang, Yinan Zheng, Grace Yoon, Zhou Zhang, Tao Gao, Brian Joyce, Wei Zhang, Joel Schwartz, Pantel Vokonas, Elena Colicino, Andrea Baccarelli, Lifang Hou, Lei Liu

{"title":"Regularized estimation in sparse high-dimensional multivariate regression, with application to a DNA methylation study.","authors":"Haixiang Zhang, Yinan Zheng, Grace Yoon, Zhou Zhang, Tao Gao, Brian Joyce, Wei Zhang, Joel Schwartz, Pantel Vokonas, Elena Colicino, Andrea Baccarelli, Lifang Hou, Lei Liu","doi":"10.1515/sagmb-2016-0073","DOIUrl":null,"url":null,"abstract":"<p><p>In this article, we consider variable selection for correlated high dimensional DNA methylation markers as multivariate outcomes. A novel weighted square-root LASSO procedure is proposed to estimate the regression coefficient matrix. A key feature of this method is tuning-insensitivity, which greatly simplifies the computation by obviating cross validation for penalty parameter selection. A precision matrix obtained via the constrained ℓ1 minimization method is used to account for the within-subject correlation among multivariate outcomes. Oracle inequalities of the regularized estimators are derived. The performance of our proposed method is illustrated via extensive simulation studies. We apply our method to study the relation between smoking and high dimensional DNA methylation markers in the Normative Aging Study (NAS).</p>","PeriodicalId":48980,"journal":{"name":"Statistical Applications in Genetics and Molecular Biology","volume":null,"pages":null},"PeriodicalIF":0.8000,"publicationDate":"2017-07-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1515/sagmb-2016-0073","citationCount":"5","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Statistical Applications in Genetics and Molecular Biology","FirstCategoryId":"100","ListUrlMain":"https://doi.org/10.1515/sagmb-2016-0073","RegionNum":4,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"BIOCHEMISTRY & MOLECULAR BIOLOGY","Score":null,"Total":0}

引用次数: 5

Abstract

In this article, we consider variable selection for correlated high dimensional DNA methylation markers as multivariate outcomes. A novel weighted square-root LASSO procedure is proposed to estimate the regression coefficient matrix. A key feature of this method is tuning-insensitivity, which greatly simplifies the computation by obviating cross validation for penalty parameter selection. A precision matrix obtained via the constrained ℓ1 minimization method is used to account for the within-subject correlation among multivariate outcomes. Oracle inequalities of the regularized estimators are derived. The performance of our proposed method is illustrated via extensive simulation studies. We apply our method to study the relation between smoking and high dimensional DNA methylation markers in the Normative Aging Study (NAS).

Abstract Image

查看原文本刊更多论文

稀疏高维多元回归中的正则化估计，并应用于DNA甲基化研究。

在本文中，我们考虑相关高维DNA甲基化标记的变量选择作为多变量结果。提出了一种新的加权平方根LASSO方法来估计回归系数矩阵。该方法的一个关键特点是调优不敏感，避免了惩罚参数选择的交叉验证，大大简化了计算。通过约束最小化方法得到的精度矩阵用于解释多变量结果之间的主体内相关性。推导了正则估计量的Oracle不等式。我们提出的方法的性能是通过广泛的仿真研究说明。我们应用我们的方法在规范衰老研究(NAS)中研究吸烟与高维DNA甲基化标记之间的关系。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Statistical Applications in Genetics and Molecular Biology BIOCHEMISTRY & MOLECULAR BIOLOGY-MATHEMATICAL & COMPUTATIONAL BIOLOGY

自引率

11.10%

发文量

期刊介绍： Statistical Applications in Genetics and Molecular Biology seeks to publish significant research on the application of statistical ideas to problems arising from computational biology. The focus of the papers should be on the relevant statistical issues but should contain a succinct description of the relevant biological problem being considered. The range of topics is wide and will include topics such as linkage mapping, association studies, gene finding and sequence alignment, protein structure prediction, design and analysis of microarray data, molecular evolution and phylogenetic trees, DNA topology, and data base search strategies. Both original research and review articles will be warmly received.