Identification of novel blood-based biomarkers of Alzheimer's Disease by pseudotime analysis

通过伪时间分析鉴定阿尔茨海默病的新型血液生物标志物

• 批准号: 10431743
• 负责人: Ignazio Stefano Piras
• 金额: $ 19.2万
• 依托单位: TRANSLATIONAL GENOMICS RESEARCH INST
• 依托单位国家: 美国
• 财政年份: 2022
• 资助国家: 美国
• 起止时间: 2022-07-01 至 2024-06-30
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/10431743
• 关键词: Algorithms; Alzheimer' s Disease; Alzheimer' s disease diagnosis; Alzheimer' s disease patient; Alzheimer’s disease biomarker; Appearance; Autopsy; Biological Markers; Blood; Brain; Cerebrospinal Fluid; Clinical; Collection; Consumption; Correlation Studies; Cross-Sectional Studies; Data; Data Set; Dementia; Development; Diagnosis; Diet; Disease; Disease Progression; Exercise; Exhibits; Expression Profiling; Future; Gene Expression; Gene Expression Profiling; Generations; Genes; Genetic Transcription; Goals; Gold; Intervention; Light; Machine Learning; Medicine; Methods; Molecular; Monitor; Neurodegenerative Disorders; Onset of illness; Pathologic; Pathway Analysis; Pathway interactions; Pharmaceutical Preparations; Pharmacological Treatment; Phase; Physical Examination; Positron-Emission Tomography; Process; Publishing; RNA; Research; Sampling; Senile Plaques; Specimen; Staging; Symptoms; Time; Tissues; Translating; Treatment Factor; United States; Validation; Whole Blood; base; blood-based biomarker; brain tissue; candidate marker; clinical practice; cost; density; differential expression; effective therapy; follow-up; lifestyle factors; machine learning algorithm; machine learning method; minimally invasive; neuroimaging; non-demented; novel; pre-clinical; rural area; single cell sequencing; symptomatic improvement; transcriptome sequencing

PROJECT SUMMARY Alzheimer's Disease (AD) is the most prevalent neurodegenerative disease in United States. Current medications are only effective at improving the symptoms for a short period of time and blood-based biomarkers for the disease are only recently beginning to emerge in research and clinical practice. In this proposal we aim to apply pseudotime analysis on publicly available RNA profiling data to detect both novel molecular processes in brain tissue and blood-based RNA biomarkers associated with AD progression. Pseudotime algorithms are machine learning approaches capable of extracting latent temporal information to order samples along a pseudotemporal progression. These approaches utilize cross-sectional data without the need of disease stage information or longitudinal specimen sampling making them uniquely well suited to the large collection of cross-sectional gene expression data currently available for AD. In Aim 1 we will focus on post-mortem brain gene expression analysis, using RNA sequencing data from bulk sampled brain tissue as well as single cell sequencing studies (e.g., Mount Sinai, ROSMAP) that include clinical and neuropathological variables related to AD staging. After extracting the pseudotime trajectories with the phenoPath method, we will prioritize genes according to their statistical correlation with pseudotime. Molecular processes associated with disease onset and progression will be inferred by Weighted Gene Coexpression Network Analysis (WGCNA). In Aim 2 we will focus on RNA expression profiling data from whole blood. Pseudotime trajectories will be determined from existing AD patient blood-based gene expression data as in aim 1, and genes will be prioritized according to their correlation with pseudotime. Then, we will retain genes highly correlated with pseudotime that simultaneously exhibit significant differential expression when compared to control samples, with the goal of finding genes that demonstrate a gradient of expression change from a non-pathological to a pathological stage that are also correlated with disease progression. Finally, we will validate the findings obtained from whole blood in post-mortem brain data from Aim 1, to assess the correlation with the gold- standard neuropathological-based staging. The findings from this proposal will allow us to identify targets for new AD treatments and identify potential candidate blood-based biomarkers of AD progression.

项目摘要 阿尔茨海默氏病（AD）是美国最普遍的神经退行性疾病。当前的 药物仅在短时间内有效改善症状和基于血液的症状 该疾病的生物标志物直到最近才在研究和临床实践中出现。 在此提案中，我们旨在将伪分析应用于公开可用的RNA分析数据以检测 与AD进展相关的脑组织和基于血液的RNA生物标志物中的新分子过程。 伪段算法是能够将潜在时间信息提取到的机器学习方法 沿伪颞进程的订单样品。这些方法利用横截面数据没有 需要疾病阶段信息或纵向标本采样，使其非常适合于 当前可用于AD的大量横截面基因表达数据。 在AIM 1中，我们将使用RNA测序数据的数据，以此 批量抽样脑组织以及单细胞测序研究（例如，西奈山，rosmap）包括 临床和神经病理学变量与AD分期有关。在提取伪段轨迹之后 该苯酚法，我们将根据基因与假频率的统计相关性优先考虑其优先级。 与疾病发作和进展相关的分子过程将通过加权基因推断 共表达网络分析（WGCNA）。 在AIM 2中，我们将重点关注来自全血的RNA表达分析数据。伪段轨迹将是 由AIM 1中的现有AD患者血液基因表达数据确定，基因将是 根据它们与假期的相关性优先级。然后，我们将保留与高度相关的基因 与对照样品相比 目的是找到证明表达梯度从非病理到一个的基因的变化 病理阶段也与疾病进展相关。最后，我们将验证发现 从AIM 1的验尸大脑数据中从全血中获得，以评估与金的相关性 标准神经病理学分期。该提案的发现将使我们能够确定 新的AD处理，并确定潜在的候选血液进展生物标志物。