Multivariate analyses of polygenic risk for impulsivity and alcohol use: A study of longitudinal trajectories of binge drinking in emerging adulthood

(7)项目总结/摘要 长期目标：该应用程序的总体目标是（1）利用先进的多元 全基因组关联研究（GWAS）的方法，以改善目前的模型遗传风险的冲动， （2）应用改进的遗传易感性模型和多基因风险评分技术预测 改变与发育相关的纵向样本中的酗酒行为。申请人的主要 我的职业目标是发展一个研究项目，整合复杂的统计遗传学建模 方法与经验知情的个性，神经认知和成瘾科学，以调查遗传 对问题性饮酒和相关疾病的发展病因学的影响。 具体目标：拟议的项目旨在（1a）计算基于单核苷酸多态性的遗传力 估计和冲动和酒精使用表型的遗传相关性，（1b）进行探索性和 冲动和饮酒表型的验证性多变量GWAS分析，以及（2）利用总结 从多变量GWAS方法中获得的统计数据，用于计算多基因风险评分以预测狂欢 在青春期后期和成年早期饮酒。为了完成拟议项目，申请人 将接受高级统计方法的广泛培训，以模拟遗传和表型数据， 数量和分子遗传学以及酒精使用和外化障碍病因学领域的专家。 培训将通过（1）课程作业，（2）参加会议和研讨会，以及（3）与 在冲动和酗酒以及统计遗传学和纵向建模方面的专家顾问。 方法：为了达到上述目的，申请人将要求并获得相关的冲动， 酒精使用GWAS汇总统计（发现样本;见研究策略部分的表1）。 来自三个独立纵向样本的全基因组遗传和表型酗酒数据， 由Slutske博士（共同申办者）和Fromme博士（顾问）提供的新生儿成人（目标样本）。 在获得发现样本数据后，将采用连锁不平衡评分回归， 生成基于单核苷酸多态性的遗传力的性状内估计值和性状间估计值 基因相关性的证据接下来，将使用两个高级多变量GWAS分析汇总统计量 结合遗传相关性以产生优化的性状特异性和潜在双系统的方法 冲动性汇总统计。最后，这些统计量将用于多基因模型中，以预测纵向 在每个目标样本中，青少年晚期和成年早期的酗酒变化。 意义：该项目的结果将增加对导致变化的遗传因素的理解 在酗酒的关键发展时期，更广泛地说，酒精使用的遗传病因学 无序发展这些发现也将代表最先进的遗传建模的新应用。

(7) Project Summary/Abstract Long Term Objectives: The overarching goals of this application are to (1) utilize advanced multivariate genome-wide association study (GWAS) approaches to improve current models of genetic risk for impulsivity, and (2) apply improved genetic liability models and polygenic risk scoring techniques to the prediction of changing binge drinking behaviors in developmentally relevant longitudinal samples. The applicant’s main career objective is to develop a program of research integrating sophisticated statistical genetics modeling approaches with empirically informed personality, neurocognitive and addictions science to investigate genetic influences on the developmental etiology of problematic drinking and related disorders. Specific Aims: The proposed project aims to (1a) calculate single nucleotide polymorphism-based heritability estimates and genetic correlations for impulsivity and alcohol use phenotypes, (1b) conduct exploratory and confirmatory multivariate GWAS analyses of impulsivity and alcohol use phenotypes, and (2) utilize summary statistics obtained from multivariate GWAS approaches to calculate polygenic risk scores to predict binge drinking across late adolescence and early adulthood. In order to complete the proposed project, the applicant will receive extensive training in advanced statistical approaches to model genetic and phenotypic data from experts in the fields of quantitative and molecular genetics and alcohol use and externalizing disorder etiology. Training will be obtained via (1) coursework, (2) conference and workshop attendance, and (3) meetings with expert consultants in impulsivity and binge drinking as well as statistical genetics and longitudinal modeling. Method: Towards the abovementioned aims, the applicant will request and acquire relevant impulsivity and alcohol use GWAS summary statistics (discovery samples; see Table 1 in Research Strategy section). Genome-wide genetic and phenotypic binge drinking data from three independent longitudinal samples of emerging adults (target samples) provided by Dr. Slutske (Co-Sponsor) and Dr. Fromme (consultant). Following discovery sample data acquisitions, linkage disequilibrium score regression will be employed to generate within-trait estimates of single nucleotide polymorphism-based heritability and between-trait estimates of genetic correlations. Next, summary statistics will be analyzed using two advanced multivariate GWAS approaches that incorporate genetic correlations to generate optimized trait-specific and latent dual-systems impulsivity summary statistics. Finally, these statistics will be used in polygenic models to predict longitudinal changes in binge drinking across late adolescence and early adulthood in each target sample. Significance: Results from this project will increase understanding of genetic factors contributing to changes in binge drinking across a critical developmental period, and more broadly, the genetic etiology of alcohol use disorder development. Findings will also represent novel applications of state-of-the-art genetic modeling.