Computational Methods for Identification of Genetic Factors Affecting the Response to Drug Abuse

项目摘要/摘要 由于药物滥用的发病率和社会成本的增加，确定影响药物滥用的遗传因素 对滥用药物(DOA)的反应特别令人感兴趣，因为这将有助于识别高危人群 并可能为治疗开发提供潜在的新靶点。然而，生物医学领域的一个主要挑战 科学正在确定一个种群内的遗传差异如何影响特性(即表型、性状) 一个人的。使用传统方法，往往需要多年的艰苦工作才能发现和 描述影响给定表型反应的遗传变异的特征。几年前，我们开发了一种更 一种将基因定位到性状的有效方法，称为基于单倍型的计算遗传图谱(HBCGM)。 在HBCGM实验中，对近交系小鼠的一种感兴趣的特性进行了测量；遗传因素是 通过识别与遗传变异模式相关的基因组区域进行计算预测 与品系之间的性状值分布有关。HBCGM分析的完成速度比 常规的遗传分析方法。然而，用于遗传因素实验验证的方法 有局限性，而且很耗时。 该项目将进一步开发计算方法，使遗传因素能够影响许多重要的 生物医学特征有待发现和实验表征。HBCGM的高通量版本(HT- HBCGM)将用于分析8,225个公开可用的数据集，这些数据集在小组中衡量213,000个回复 近亲繁殖的小鼠品系。我们部署了一种新的方法，通过利用 在检查相似反应的许多数据集中存在冗余。新的计算工具， 还将促进对遗传、转录和代谢组数据的综合分析。这 包括专门的新陈代谢网络(用于大脑和其他3个组织)，用于计算识别 与基因表达或遗传差异相关的代谢变化。为了激励其他调查人员 为了进行基因发现，这个项目的所有结果和方法都将完全提供给科学家 社区。这些计算工具将被用来分析定制的“多体”(遗传、转录、 和代谢)数据集，其测量：(I)近交系菌株组对四个DOA(可卡因， 甲基苯丙胺、芬太尼和尼古丁)；和(Ii)相应的DOA诱导转录和 脑内代谢的变化。对这些数据的综合分析将确定影响 对死亡时间的反应。然后，我们应用一种高效的方法将特定的等位基因改变工程到 自交系的基因组和工程化品系被用来实验测试已鉴定的 遗传因素对DOA的反应。

Project Summary/Abstract Due to the increased morbidity and societal cost of drug abuse, identification of genetic factors affecting the response to drugs of abuse (DOA) are of particular interest because this will aid in identifying at risk populations and could provide potential novel targets for therapeutic development. However, a major challenge in biomedical science is determining how genetic differences within a population affect the properties (i.e. phenotypes, traits) of an individual. Using conventional methods, it often requires years of painstaking work to discover and characterize a genetic variant that affects a given phenotypic response. Several years ago, we developed a more efficient method for mapping genes to traits, called haplotype-based computational genetic mapping (HBCGM). In an HBCGM experiment, a property of interest is measured in inbred mouse strains; and genetic factors are computationally predicted by identifying the genomic regions where the pattern of genetic variation correlates with the distribution of trait values among the strains. HBCGM analyses are completed much more quickly than conventional genetic analysis methods. However, the methods used for experimental validation of genetic factors have limitations and are time consuming. This project will further develop computational methods that will enable genetic factors affecting many important biomedical traits to be discovered and experimentally characterized. A high-throughput version of HBCGM (HT- HBCGM) will be used to analyze 8,225 publicly available datasets, which measure 213,000 responses in panels of inbred mouse strains. We deploy a novel method that increases genetic discovery power by exploiting the redundancy present in the many datasets that examine similar responses. Novel computational tools that facilitate the integrated analysis of genetic, transcriptional and metabolomic data will also be developed. This includes specialized metabolic networks (for brain and 3 other tissues) for computationally identifying metabolomic changes that correlate with gene expression or genetic differences. To stimulate other investigators to make genetic discoveries, all results and methods from this project will be made fully available to the scientific community. These computational tools will be used to analyze customized ‘multi-omic’ (genetic, transcriptional, and metabolomic) datasets that measure: (i) fifteen responses of inbred strain panels to four DOA (cocaine, methamphetamine, fentanyl, and nicotine); and (ii) corresponding DOA induced transcriptional and metabolomic changes in brain. Integrated analysis of this data will identify genes/pathways affecting the response to DOA. We then apply a high efficiency method for engineering specific allelic changes into the genome of inbred strains, and the engineered lines are used to experimentally test the effect of an identified genetic factor on the response to a DOA.