Understanding Selectivity Mechanisms of Network Vulnerability and Resilience in Alzheimer's Disease by Establishing a Neurobiological Basis through Network Neuroscience

抽象的。大量的神经科学和神经影像学研究表明，阿尔茨海默病（AD） 不同地影响大脑的某些区域和特定的细胞类型。由于AD相关病理事件 通常跨神经传播，对神经元损失和结构损伤的选择性脆弱性也表现出来。 网络变化的拓扑模式。沿着许多其他研究，研究小组发现， 强有力的证据表明，（1）AD优先影响网络中密集连接的枢纽节点， 网络，和（2）神经病理学负担的传播，如淀粉样斑块和神经胶质瘤 缠结表现出由自组织谐波基底控制的独特拓扑图案。然而，在这方面， 这种网络脆弱性的潜在因素和调节AD选择性的分子机制 仍然不清楚。在这方面，我们的目标是继续开发尖端的网络分析工具， 对神经病理学事件如何选择性地影响某些谐波的更深入的方法论理解 基础（谐波选择性网络脆弱性）以及大脑网络如何对抗AD病理（网络 弹性）。在这种情况下，该项目的骨干是一个谐波因子分析模型，可用作 神经生物学基础，以准确表征系统中神经变性的全脑映射 水平，其中每个谐波因子解释了神经病理学的普遍传播（波）模式 事件从特定的结构连接体途径出现。在目标1中，我们将充分利用 功率和能量的生物物理学概念，以确定一组谐波选择性脆弱模式， 正常老化和AD之间的网络脆弱性。此外，我们将关联已识别的网络漏洞 与来自不同研究领域的耦合因素，包括选择性随机（统计），系统 关键性（物理学），网络组织（网络神经科学）和认知领域（临床）。之后我们 将根据发现的网络之间的关联寻找推定的和谐遗传学生物标志物， 目标2中的脆弱性和遗传因素，并开发一种和谐遗传方法来捕获网络弹性 目标3。在目标4中，我们将应用目标1 - 3中开发的计算方法来建立（1）一个精细的 对A（淀粉样蛋白-PET）、T（Tau-PET）和N（FDG-PET）网络脆弱性和弹性的全面理解。 PET和皮质厚度）生物标志物，以及（2）网络动态的纵向基础 通过研究AT [N]生物标志物的纵向变化来评估脆弱性。我们小说的诊断能力 和谐遗传学生物标志物和弹性将在我们目前的AD诊断引擎中进行评估。我们将 发布软件（二进制程序和源代码），以促进其他AD生物标志物项目， 与脑网络功能障碍相关的其他神经认知障碍的神经影像学研究。

Abstract. A plethora of neuroscience and neuroimaging studies have shown that Alzheimer’s disease (AD) differentially affects certain regions of the brain and specific cell types. Since AD-related pathological events often propagate trans-neuronally, the selective vulnerability to neuron loss and structure damage also manifest in the topological patterns of network alteration. Along with many other studies, the research team has found the strong evidence that (1) AD preferentially affects hub nodes in the network that are densely connected in the network, and (2) the propagation of neuropathological burdens such as amyloid plaques and neurofibrillary tangles exhibit unique topological patterns that are governed by the self-organized harmonic bases. However, the factors underlying this network vulnerability and the molecular mechanism regulating the selectivity in AD remain unclear. In this regard, we aim to continue the development of cutting-edge network analysis tools with a greater methodological understanding of how neuropathological events selectively affect certain harmonic bases (harmonic-selective network vulnerability) and how brain networks counteract AD pathology (network resilience). In this context, the backbone of this project is a harmonic factor analysis model that can be used as a neurobiological basis to accurately characterize the whole-brain mapping of neurodegeneration at a system level, where each harmonic factor explains how the ubiquitous propagation (wave) pattern of neuropathological event emerges from the particular structural connectome pathway. In Aim 1, we will leverage the well-studied biophysics concept of power and energy to identify a set of harmonic-selective vulnerable patterns that account for network vulnerability between normal aging and AD. Also, we will associatethe identified network vulnerability with couple factors from diverse research fields which include stochastics of selectivity (statistics), system criticality (physics), network organization (network neuroscience), and cognitive domains (clinic). After that, we will seek for the putative harmonic-genetics biomarker based on the discovered association between network vulnerability and genetics factor in Aim 2 and develop a harmonic-genetic approach to capture network resilience in Aim 3. In Aim 4, we will apply the computational approaches developed in Aim 1-3 to establish (1) a fine- grain understanding of network vulnerability and resilience across A (amyloid-PET), T (Tau-PET), and N (FDG- PET and cortical thickness) biomarkers, and (2) a longitudinal underpinning of the dynamics of network vulnerability by investigating the longitudinal change of AT[N] biomarkers. The diagnostic power of our novel harmonic-genetics biomarker and resilience will be evaluated in our current AD diagnostic engines. We will release the software (both binary program and source code), to facilitate the other AD biomarker projects and the neuroimaging studies of other neurocognitive disorders associated with brain network dysfunction.