Studying the Genetics of Aging, Behavioral, and Social Phenotypes in Diverse Populations

研究不同人群的衰老、行为和社会表型的遗传学

• 批准号: 10638152
• 负责人: Patrick Ansel Turley
• 金额: $ 72.82万
• 依托单位: UNIVERSITY OF SOUTHERN CALIFORNIA
• 依托单位国家: 美国
• 财政年份: 2023
• 资助国家: 美国
• 起止时间: 2023-05-15 至 2028-04-30
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10638152
• 关键词: Address; African; Aging; Algorithms; Alzheimer' s Disease; American; Behavioral; Characteristics; China; Communication; Computer software; Data; Data Set; Dedications; Dimensions; Distant; Documentation; East Asian; Education; Environment; Ethics; European; European ancestry; General Population; Genes; Genetic; Genetic Research; Genetic study; Genomics; Genotype; Grant; Health; Human Resources; Individual; Knowledge; Linkage Disequilibrium; Methods; Modeling; Outcome; Paper; Performance; Persons; Phenotype; Population; Population Genetics; Population Heterogeneity; Publications; Publishing; Research; Research Personnel; Sample Size; Sampling; Siblings; Social Sciences; Source; South Asian; Statistical Bias; Weight; Writing; behavioral phenotyping; biobank; cohort; computer studies; computerized tools; cost; design; diverse data; ethical, legal, and social implication; experience; genetic association; genome wide association study; genome-wide; improved; indexing; novel strategies; online repository; prevent; repository; simulation; social; statistics; tool; trait

Project Summary/Abstract For this application, “Studying the Genetics of Aging, Behavioral, and Social Phenotypes in Diverse Populations,” we propose to develop tools to promote genetic research of aging, behavioral, and social phenotypes in diverse populations. These phenotypes have a number of unique characteristics (e.g., polygenicity, environmental mechanisms, and small effect sizes) which require special consideration when developing research tools. In brief, we propose to: • Develop the Genetic-Related-Matrix-Matched Association study (GRMMA) tool for performing genome-wide association studies (GWASs) in large, diverse data sets. Current GWAS methods require restricting samples into approximately homogeneous-ancestry samples, which is wasteful and has resulted in Eurocentric bias in genetics research. Using matching methods, GRMMA can use more of the available data in a way that both reduces bias and increases statistical power. We will employ computationally efficient strategies that allow us to implement GRMMA in large diverse sample such as the UK Biobank. We will make the GRMMA tool and tutorials publicly available through the online repository, Github. • Develop SBayes-Universal (SBayesU), an efficient new tool for producing polygenic scores (PGSs) by optimally combining GWAS summary statistics estimated in different populations. The key feature of SBayesU is that it uses a low-dimensional eigen decomposition of the linkage disequilibrium matrix. This permits SBayesU to model a much larger set of SNPs, to model SNP annotations, to account for imperfect cross-ancestry genetic correlation, to produce PGSs for populations that are not included among the sets of GWAS summary statistics, and to allow our algorithms to converge much more quickly and reliably. We will also make the SBayesU tool and tutorials publicly available. • We will apply the best available method for producing diverse-population PGSs (which we anticipate will be SBayesU) to a wide range of aging, behavioral, and social phenotypes, using existing cohorts and new genotyped data that becomes available during the grant period. We will make the polygenic scores we produce publicly available as part of the Social Science Genetic Association Consortium’s Polygenic Index Repository, which currently creates polygenic scores for 11 widely used datasets (but currently only for the European-ancestry individuals in those datasets). Each release of the Repository will be accompanied by documentation that clearly describes methods used and the underlying data.

项目概要/摘要 对于此应用程序，“研究不同人群中衰老、行为和社会表型的遗传学”， 我们建议开发工具来促进衰老、行为和社会表型的基因研究 这些表型具有许多独特的特征（例如，多基因性， 环境机制和小效应量），在开发时需要特别考虑 简而言之，我们建议： • 开发遗传相关基质匹配关联研究 (GRMMA) 工具来执行 当前的 GWAS 方法需要在大型、多样化的数据集中进行全基因组关联研究 (GWAS)。 将样本限制为近似同源的样本，这是浪费的并且有 导致遗传学研究中的欧洲中心偏差。使用匹配方法，GRMMA 可以使用更多的 我们将采用既减少偏差又提高统计功效的可用数据。 计算高效的策略使我们能够在大量不同的样本中实施 GRMMA，例如 英国生物银行。我们将通过在线公开提供 GRMMA 工具和教程。 存储库，Github。 • 开发 SBayes-Universal (SBayesU)，这是一种通过以下方式生成多基因评分 (PGS) 的高效新工具： 最佳结合不同人群估计的 GWAS 汇总统计数据的关键特征。 SBayesU的特点是它使用了连锁不平衡矩阵的低维特征分解。 这使得 SBayesU 能够对更大的 SNP 集进行建模，对 SNP 注释进行建模，以解释 不完美的跨血统遗传相关性，为未包括在内的人群生成 PGS 在 GWAS 汇总统计数据集中，让我们的算法更加收敛 我们还将快速可靠地公开 SBayesU 工具和教程。 • 我们将采用最佳可用方法来生产不同群体的 PGS（我们预计 将是 SBayesU），利用现有的队列和 我们将在资助期间提供新的基因分型数据。 我们作为社会科学遗传学协会联盟多基因的一部分公开提供 索引存储库，目前为 11 个广泛使用的数据集创建多基因分数（但目前 仅适用于这些数据集中的欧洲血统个体）。 附有清楚描述所用方法和基础数据的文档。