Data Analysis and Management

数据管理和分析核心（核心D）将在研究设计和数据管理与分析的各个方面为项目和其他核心提供支持。该POI侧重于通过成像技术表征特定肥胖表型，识别生活方式和生物标志物变量中的表型预测因子，这些变量来源于各种生物学维度（遗传学、代谢组学、微生物组学）的测量，并测试预测因子（作为替代测量）与 几种癌症的风险。研究受试者是成熟的多种族队列（MEC）研究的参与者。核心D的经验丰富的生物统计学家和生物信息学家共同拥有MEC数据和设计以及支持P01目标所需的数据和分析类型的丰富经验。具体目标包括（1）加强复杂的MEC信息系统，以确定招募的队列成员，跟踪他们在研究中的进展，并监测接收情况， 所有生物材料的储存和运输，（2）预处理来自遗传变异、代谢组学和肠道微生物组的高通量数据，以减少维度并提供最合适的分析数据，以及（3）与项目负责人一起为P01申请的每个研究项目制定和执行分析计划。特别是，将使用适当的统计工具和软件来寻找研究中肥胖表型的最具预测性的模型，并将这些预测因子与癌症风险进行关联研究。将在预测和关联模型中检查和测试各种族的异质性。

The Data Management and Analysis Core (Core D) will provide support for the projects and other cores in all aspects of study design and data management and analysis. This POl focuses on the characterization of specific adiposity phenotypes by imaging techniques, the identification of predictors of the phenotypes among lifestyle and biomarker variables derived from measurements in various biological dimensions (genefics, metabolomics, microbiome), and testing the association of the predictors, as proxy measures, with the risk of several cancers. The study subjects are participants in the well-established Multiethnic Cohort (MEC) study. The experienced biostatisticians and bioinformaticians of Core D collectively have extensive experience with the MEC data and design, and the types of data and analyses needed to support the aims of the P01. Specific aims include (1) to enhance the complex MEC informatics system to identify cohort members to approach for recruitment, to track their progress through the study, and to monitor receipt, storage and shipment of all biologic materials, (2) to pre-process high through-put data from genetic variation, metabolomics and gut microbiome in order to reduce the dimensionality and provide the most appropriate data for analysis, and (3) to develop and execute analysis plans for each of the research projects of this P01 application, in conjunction with the Project Leaders. In particular, appropriate statistical tools and software will be used to find the most predictive models of the adiposity phenotype under study and for the association studies of these predictors with cancer risk. Heterogeneity will be examined and tested across the ethnic groups in the prediction and association models.