Detecting synergistic effects of pharmacological and non-pharmacological interventions for AD/ADRD

项目摘要 阿尔茨海默病（AD）及其相关痴呆（AD/ADRD）是一种多因素异质性疾病。最 目前的研究集中在药物干预AD/ADRD，目前没有药物干预（PI） 发现性研究已经转化为有效的治疗方法。然而，越来越多的证据表明， 非药物干预（NPI），如睡眠，饮食，膳食补充剂，有氧运动， 芳香疗法，光疗法，认知训练，是潜在的修改，从而提供替代的机会 AD/ADRD预防。因此，本项目的目标是开发翻译信息学方法， 使用多模式汇总、标准化和发现药物和NPI候选物对AD/ADRD的影响 数据资源（即，生物医学文献、EHR、临床试验），然后进行动物模型验证。实现 我们的目标是，我们提出以下目标：1）构建全面的药理学和非药理学 来自生物医学文献的阿尔茨海默病药物干预知识图谱（PANIA-KG） 和其他知识库; 2）检测、理解和可视化PI的药物再利用信号， NPI及其对AD/ADRD的协同作用（使用PANIA-KG）; 3）重新排序和验证个体 以及使用多模态数据源和动物模型的协同药物再利用信号。成功 该项目的完成将提供全面的NPI知识图，新颖的信息学方法， 药物和NPI候选药物的排名列表，并使用多模式数据源验证协同干预。 生成的方法、PANIA-KG和排序列表可以进一步促进我们的临床研究和临床试验 该设计关注药物和NPI对AD/ADRD的协同作用。

PROJECT SUMMARY Alzheimer's disease (AD) and related dementia (AD/ADRD) is a multifactorial and heterogeneous disorder. Most current studies focus on drug interventions for AD/ADRD and currently none of pharmacological intervention (PI) discovery research has been translated into effective treatments. However, increasing evidence demonstrates that non-pharmacological interventions (NPI), such as sleep, diet, dietary supplements, aerobic exercise, aromatherapy, light therapy, cognitive training, are potentially modifiable and thus offer alternative opportunities for AD/ADRD prevention. Thus, the objective of this project is to develop translational informatics approaches to aggregate, standardize and discover the effects of drug and NPI candidates on AD/ADRD using multi-modal data resources (i.e., biomedical literature, EHR, clinical trials) followed by animal model validation. To achieve our goal, we propose the following aims: 1) constructing a comprehensive Pharmacological And Non- Pharmacological Interventions for Alzheimer's Disease Knowledge Graph (PANIA-KG) from biomedical literature and other knowledge bases; 2) detecting, understanding, and visualization of drug repurposing signals of PIs, NPIs, and their synergistical effects for AD/ADRD using the PANIA-KG; 3) re-ranking and validating individual and synergistical drug repurposing signals using multimodal data sources and animal models. The successful completion of this project will deliver a comprehensive NPI knowledge graph, novel informatics approaches, ranked list of drug and NPI candidates, and validated synergistic intervention using multi-modal data sources. The generated approaches, PANIA-KG and ranked lists can further our clinical investigations and clinical trial design which focuses on synergistic effects of drug and NPIs for AD/ADRD.