High-Dimensional Data Issues in Aging Research

衰老研究中的高维数据问题

• 批准号: 7862921
• 负责人: Bin Nan
• 金额: $ 28.15万
• 依托单位: UNIVERSITY OF MICHIGAN AT ANN ARBOR
• 依托单位国家: 美国
• 财政年份: 2010
• 资助国家: 美国
• 起止时间: 2010-04-15 至 2014-03-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/7862921
• 关键词: Address; Age related macular degeneration; Aging; Algorithms; Alzheimer' s Disease; Amyloid deposition; Biomedical Research; Brain; Brain imaging; Collaborations; Computer software; DNA Methylation; Data; Data Analyses; Dementia; Dependence; Development; Dimensions; Disease; Epigenetic Process; Genetic; Genome; Genomics; Goals; Group Structure; Health Sciences; Image; Impaired cognition; Individual; Inequality; Linear Models; Linear Regressions; Literature; Maps; Measures; Methodology; Methods; Michigan; Modeling; Motor; Nerve Degeneration; Neurodegenerative Disorders; Parkinson Disease; Patients; Performance; Pittsburgh Compound-B; Positron; Positron-Emission Tomography; Principal Component Analysis; Procedures; Property; Public Health; Radio; Research; Research Personnel; Sampling; Severities; Signal Transduction; Staging; Structure; Testing; Tracer; Universities; age related; aging population; base; case control; cognitive function; cohort; computer program; disease diagnosis; follow-up; gene discovery; genome wide association study; health science research; interest; markov model; novel; programs; public health relevance; research and development; response; simulation; statistical center; statistics; theories; tomography; user-friendly

DESCRIPTION (provided by applicant): The broad objectives of this research are developments of regularization methods for high-dimensional data that arise commonly in biomedical studies, particularly studies in genomics, epigenetics, and brain imaging. The specific aims in this proposal are motivated by problems arising in studies of neurodegenerative disorders in an aging population including epigenetic and PiB PET studies for patients with Alzheimer's disease, PET studies of dementia in patients with Parkinson's disease, and the genome-wise association study for the age- related macular degeneration. They include: 1. Developing effective variable selection methods in high- dimensional regression models with grouped structures. We will consider the regularization with a novel convex penalty in multiple linear regression with high-dimensional grouped covariates, generalized linear models with high-dimensional grouped covariates, and multivariate linear regression with both high- dimensional grouped response variables and high-dimensional grouped covariates. Fast algorithms will be developed, theoretical properties such as oracle inequalities will be examined, and finite sample performance will be evaluated through extensive simulations. 2. Developing and evaluating regularization methods for disease diagnosis with functional data. We consider functional linear regression models and functional principal component analysis for imaging data to achieve sparse group effects. Both theoretical and numerical performance of the proposed methods will be examined. 3. Developing new multiple testing methodologies for dependent tests. We propose to use the hidden Markov models, either non-homogeneous or group homogeneous, to characterize the dependence structure among the multiple tests, and develop procedures with enhanced power and correctly controlled false discovery rate. We will examine the asymptotic optimality and numerical performance of the proposed methods. 4. Developing user-friendly computing programs systematically for the proposed statistical methodologies and disseminating them to health sciences researchers. PUBLIC HEALTH RELEVANCE: The application is intended to solve emerging statistical issues in high-dimensional data analysis, which arise commonly in biomedical and public health studies. The proposed methods will be particularly useful in epigenetic, genomic, PET imaging studies on neurodegenerative disorders including Alzheimer's disease, Parkinson's disease, and age-related macular degeneration.

描述（由申请人提供）：本研究的广泛目标是开发生物医学研究中常见的高维数据的正则化方法，特别是基因组学，表观遗传学和脑成像研究。本提案中的具体目标是由老龄化人群中神经退行性疾病研究中出现的问题激发的，包括阿尔茨海默病患者的表观遗传和PiB PET研究、帕金森病患者痴呆的PET研究以及年龄相关性黄斑变性的基因组关联研究。它们包括：1.在分组结构的高维回归模型中发展有效的变量选择方法.我们将考虑在高维分组协变量的多元线性回归、高维分组协变量的广义线性模型以及高维分组响应变量和高维分组协变量的多元线性回归中使用新型凸罚的正则化。快速算法将被开发，理论属性，如甲骨文不等式将被检查，有限样本的性能将通过广泛的模拟进行评估。2.用函数数据开发和评估疾病诊断的正则化方法。我们认为功能线性回归模型和功能主成分分析成像数据，以实现稀疏组的影响。所提出的方法的理论和数值性能进行检查。3.为相关测试开发新的多重测试方法。我们建议使用隐马尔可夫模型，无论是非齐次或组齐次，多个测试之间的依赖结构的特征，并开发程序，增强了电源和正确控制的错误发现率。我们将研究所提出的方法的渐近最优性和数值性能。4.为拟议的统计方法系统地开发用户友好的计算程序，并将其传播给健康科学研究人员。 公共卫生相关性：该应用程序旨在解决高维数据分析中出现的统计问题，这些问题通常出现在生物医学和公共卫生研究中。所提出的方法将是特别有用的表观遗传，基因组，PET成像研究神经退行性疾病，包括阿尔茨海默氏病，帕金森氏病，和年龄相关性黄斑变性。