Characterizing Genetic, Neurotransmitter receptors and Macroscopic Brain Interactions via Multi-scale Analytical Modeling

在我们试图理解大脑时面临的多重挑战中，我们目前在将多尺度数据(分子、神经成像、认知、行为)集成到统一的生成模型中的能力尤为突出。要解决这一问题，需要发展新的数学、神经科学和计算技术，这些技术构成了这一提议的基础。从分子(如基因、神经递质受体)到宏观(如神经成像)尺度，大脑构成了一个复杂的动态系统。众所周知，每一种认知/行为状态都是多个生物尺度之间同时存在的因果作用的结果。然而，大多数脑动力学的生成性模型只关注宏观的脑功能信号，没有刻画脑多因素动力学的分子基础，更重要的是，没有考虑其他关键的生物因素(如脑血管流量、代谢、神经元组织密度)。本研究的目的是：i)构建一个综合的数学模型，以考虑基因、神经递质受体、多因素宏观脑因素(神经元活动完整性、葡萄糖代谢、血管流量、灰质密度)和个体认知/行为属性之间的时空、多尺度和多因素的相互作用。我们将使用动态系统方程来定义一个统一的公式，该公式捕捉跨尺度生物因素和认知/行为状态之间的直接相互作用。来自大型数据库(例如，Allen Brain Atlas、研究中心Julich的受体图谱、基因表达总览)的数据将用于开发该模型并验证其适用性。我们假设，个性化转录-受体-成像模型参数的事后分析将揭示基本的生物学机制。2)使用确定的分子-神经成像-认知大脑框架来表征不同人口统计(性别、性别、教育水平)、生活方式(饮食、体育锻炼、睡眠习惯)和环境条件(地理位置、社会和社区背景)对每个人的大脑和认知/行为的分子和宏观影响。我们假设，对模型参数的事后分析将阐明不同人口、生活方式和环境条件的生物学影响。三)通过已经开发的开放获取、用户友好的软件，与研究界共享所有分析工具。共享整合的分子-神经成像-认知脑框架将加快其验证和国际使用。该计划将培训4名HQP(2名硕士和2名博士)，并加强我们对人类认知和行为背后的时空多尺度脑相互作用的理解；从而为从机械论的角度理解和调节多因素脑状态奠定关键基础。

Amongst the multiple challenges that we are faced when trying to understand the brain, our current inability at integrating multiscale data (molecular, neuroimaging, cognitive, behavioral) into a unified generative model stands out. To solve this problem requires the development of novel mathematical, neuroscience and computational techniques, which form the basis of this proposal. From the molecular (e.g. genes, neurotransmitter receptors) to the macroscopic (e.g. neuroimaging) scales, the brain constitutes a complex dynamic system. It is known that each cognitive/behavioral state is the result of concurrent causal interactions between the multiple biological scales. However, most generative models of brain dynamics only focus on macroscopic functional brain signals, not characterizing the molecular basis underlying brain multifactorial dynamics and, importantly, failing to consider other crucial biological factors (e.g. cerebrovascular flow, metabolism, neuronal tissue density). This research aims to: i)Construct an integrative mathematical model to consider the spatiotemporal multiscale and multifactorial interactions between genes, neurotransmitter receptors, multifactorial macroscopic brain factors (neuronal activity integrity, glucose metabolism, vascular flow, gray matter density) and individual cognitive/behavioral properties. We will use dynamic system equations to define a unifying formulation capturing the direct interactions among across-scales biological factors and cognitive/behavioral states. Data from large-scale databases (e.g. Allen Brain Atlas, receptors atlases from Research Center Julich, Gene Expression Omnibus) will be used for developing the model and validating its applicability. We hypothesize that the post-hoc analysis of personalized transcriptomic-receptor-imaging model parameters will reveal essential biological mechanisms. ii)Use the defined molecular-neuroimaging-cognitive brain framework to characterize the molecular and macroscopic impact of different demographic (sex, gender, educational level), life style (diet, physical exercise, sleep habits) and environmental conditions (geographic location, social and community context) on each individual's brain and cognition/behavior. We hypothesize that post-hoc analysis of model parameters will clarify the biological impact of different demographic, lifestyle and environmental conditions. iii)Share all resulting analytic tools with the research community via an already developed open-access user-friendly software. Sharing the integrated molecular-neuroimaging-cognitive brain framework will accelerate its validation and international use. This program will train 4 HQPs (2 MAs and 2 PhDs) and enhance our understanding of the spatiotemporal multiscale brain interactions underlying human cognition and behavior; thus, establishing the critical foundation for understanding and modulating multifactorial brain states from a mechanistic perspective.