Translational big data analytic approaches to advance drug repurposing for Alzheimer's disease

项目摘要 阿尔茨海默病(AD)是一种严重的公共卫生危机，目前尚无有效的治疗方法。考虑到最近的故障 许多AD临床试验，迫切需要制定有效的策略来确定新的AD靶点 疾病模型和药物再利用和开发的新候选者。我们在这里提出了一项研究 在翻译生物信息学领域开发变革性大数据分析方法的项目， 机器学习和深度学习促进AD药物再利用。我们的总目标是发展 创新的机器学习和深度学习方法以及信息学工具和管道， 利用相关生物医学领域的大数据。这些大数据包括大规模的基因，多组学， 来自里程碑式的AD研究的成像、认知和其他表型数据， 药物、蛋白质和疾病、药理扰动数据、电子健康记录数据和市场扫描 数据。我们提出的计算研究旨在开发新的翻译信息学方法 分析各种类型的分子、临床和其他相关数据，以识别个别药物或药物 具有良好疗效和毒性特征的组合可用于重新定位以对抗AD或AD- 相关痴呆症(ADRD)。为了实现我们的目标，我们有四个目标。目标1是开发基于网络的多 识别AD药物重新定位新靶点的基因和通路的组学数据集成方法 研究。目标2是开发信息学策略，通过以下方式确定有希望的候选目标的优先顺序并进行评估 研究它们与AD生物标记物和表型的关系。目标3是开发知识驱动的药物 利用网络强化和药物评分确定AD候选药物的再利用方法。目标4是 使用药理学对确定的候选药物进行优先级排序和评估，以重新用于对抗AD/ADRD 扰动、EHR和MarketScan数据。这些目标的成功实现将产生新颖的翻译 大数据分析方法和工具，以提高我们对遗传、分子和神经生物学的理解 阿尔茨海默病的机制，有助于确定新的有希望的靶点和药物的再利用，以及 最终对疾病的治疗和预防产生转化性影响。这些进步是根本性的 NIA NAPA到2025年有效治疗或预防AD/ADRD的目标。由此产生的方法和工具 预计还将影响一般的生物医学研究，并使公共卫生成果受益。

Project Summary Alzheimer’s disease (AD) is a major public health crisis with no available cure. Given recent failures of many AD clinical trials, there is an urgent need for developing effective strategies to identify new AD targets for disease modeling and new candidates for drug repurposing and development. We propose here a research project to develop transformative big data analytic approaches in the fields of translational bioinformatics, machine learning and deep learning to advance drug repurposing for AD. Our overarching goal is to develop innovative machine learning and deep learning approaches as well as informatics tools and pipelines that leverage big data in relevant biomedical domains. These big data include large-scale genetic, multi-omics, imaging, cognitive and other phenotypic data from landmark AD studies, functional interaction data among drugs, proteins and diseases, pharmacologic perturbation data, electronic health record data, and MarketScan data. Our proposed computational research is aimed at developing novel translational informatics approaches to analyze various types of molecular, clinical and other relevant data to identify individual drugs or drug combinations with favorable efficacy and toxicity profiles as candidates for repositioning against AD or AD- related dementia (ADRD). To achieve our goal, we have four Aims. Aim 1 is to develop network-based multi- omics data integration methods to identify genes and pathways as novel targets for AD drug repositioning research. Aim 2 is to develop informatics strategies to prioritize and evaluate promising candidate targets via examining their associations with AD biomarkers and phenotypes. Aim 3 is to develop knowledge-driven drug repurposing methods using network reinforcement and drug scoring to identify AD candidate drugs. Aim 4 is to prioritize and evaluate the identified candidate drugs for repurposing against AD/ADRD using pharmacologic perturbation, EHR and MarketScan data. Successful completion of these aims will produce novel translational big data analytic methods and tools to improve our understanding of the genetic, molecular and neurobiological mechanisms of AD, facilitate the identification of novel promising targets and drugs for repurposing, and ultimately have a translational impact on disease treatment and prevention. These advances are fundamental to the NIA NAPA goal of effectively treating or preventing AD/ADRD by 2025. The resulting methods and tools are also expected to impact biomedical research in general and benefit public health outcomes.